Skip to main content

Full text of "Learning Music With Synthesizers"

See other formats


© Copyright 1974 
ARP Instruments, Inc. 
Lexington, Massachusetts 

Second Edition 
All Rights Reserved 

Library of Congress Catalog Card Number 74-77567 

Printed in the United States of America 



David Friend 
Alan R. Pearlman 
Thomas D. Piggott 


Hal Leonard Publishing Corporation 



Introduction: Electronic Sound Synthesis 1 

Lesson 1: Waveforms 2 

Lesson 2: Overtones and Harmonics 10 

Lesson 3: Filters and Filtering 13 

Lesson 4: Low-Frequency Waveforms 17 

Lesson 5: Voltage Control 20 


Section 1: Basic Operational Features 25 

Section 2: Signal Sources 32 

Section 3: Signal Modifiers 60 

Section 4: The Controllers 91 


Section 1: Timbre 132 

Section 2: Melody 155 

Section 3: Harmony 164 

Section 4: Transposition 178 

Section 5: Setting Up an Electronic Music Studio 188 

Conclusion: Listening for Electronic Music 208 

Glossary 209 

Index 212 


This text is designed to serve as an introduction to electronic music 
synthesis. The experimental sections of the book are based on use of 
the ARP Odyssey synthesizer, although the theory and techniques can 
generally be applied to any synthesizer. 

Part I covers the basic theory needed for understanding sound synthe- 
sis, and many new terms are presented and explained. These terms are 
the language of synthesizers and should be understood by those wishing 
to expand their musical experiences through synthesizers. 

Part II is a "hands-on" approach to mastering the ARP Odyssey synthe- 
sizer. More terminology is presented and explained, and the experi- 
ments and exercises always relate to the theory presented in Part I of 
the text. The practical understanding of the operation of the synthe- 
sizer gained through studying this part of the text will quickly enable 
anyone to learn how to operate nearly any synthesizer. 

Part III of the text ties this newly mastered instrument into the frame- 
work of traditional musical concepts. The synthesizer is a remarkably 
versatile and flexible tool for the musician, and this part of the book 
explains how the synthesizer can be used to demonstrate and reinforce 
many basic concepts in music and sound. 

The material presented in this book will be best understood if the 
three major parts of this book are studied in the order presented. And 
although there is no substitute for practical experience with a synthe- 
sizer, it is possible to gain a good basic understanding of electronic music 
production by studying the many detailed diagrams in this book. 

The synthesizer represents a major evolutionary step in music and 
music education. It adapts so well to such a wide variety of applications 
that its usefulness is limited only by one's imagination. In this book we 
have offered the tools necessary to begin exploring and enjoying the 
new capabilities of electronic music and synthesizers. 

The authors would like to express their thanks to Dave Fredericks 
of ARP Instruments, Inc., and to Gary Meisner and Dick Peck of Hal 
Leonard Publishing Corporation for their invaluable assistance in- 
writing this book, and to Vivian Hutchins,' editor, and Margaret 
Shepherd for art direction. 

Newton, Massachusetts David Friend 

February 1974 



This text is designed to serve as an introduction to electronic music 
synthesis. The experimental sections of the book are based on use of 
the ARP Odyssey synthesizer, although the theory and techniques can 
generally be applied to any synthesizer. 

Part I covers the basic theory needed for understanding sound synthe- 
sis, and many new terms are presented and explained. These terms are 
the language of synthesizers and should be understood by those wishing 
to expand their musical experiences through synthesizers. 

Part II is a "hands-on" approach to mastering the ARP Odyssey synthe- 
sizer. More terminology is presented and explained, and the experi- 
ments and exercises always relate to the theory presented in Part I of 
the text. The practical understanding of the operation of the synthe- 
sizer gained through studying this part of the text will quickly enable 
anyone to learn how to operate nearly any synthesizer. 

Part III of the text ties this newly mastered instrument into the frame- 
work of traditional musical concepts. The synthesizer is a remarkably 
versatile and flexible tool for the musician, and this part of the book 
explains how the synthesizer can be used to demonstrate and reinforce 
many basic concepts in music and sound. 

The material presented in this book will be best understood if the 
three major parts of this book are studied in the order presented. And 
although there is no substitute for practical experience with a synthe- 
sizer, it is possible to gain a good basic understanding of electronic music 
production by studying the many detailed diagrams in this book. 

The synthesizer represents a major evolutionary step in music and 
music education. It adapts so well to such a wide variety of applications 
that its usefulness is limited only by one's imagination. In this book we 
have offered the tools necessary to begin exploring and enjoying the 
new capabilities of electronic music and synthesizers. 

The authors would like to express their thanks to Dave Fredericks 
of ARP Instruments, Inc., and to Gary Meisner and Dick Peck of Hal 
Leonard Publishing Corporation for their invaluable assistance in- 
writing this book, and to Vivian Hutchins,* editor, and Margaret 
Shepherd for art direction. 

Newton, Massachusetts David Friend 

February 1974 



This text is designed to serve as an introduction to electronic music 
synthesis. The experimental sections of the book are based on use of 
the ARP Odyssey synthesizer, although the theory and techniques can 
generally be applied to any synthesizer. 

Part I covers the basic theory needed for understanding sound synthe- 
sis, and many new terms are presented and explained. These terms are 
the language of synthesizers and should be understood by those wishing 
to expand their musical experiences through synthesizers. 

Part II is a "hands-on" approach to mastering the ARP Odyssey synthe- 
sizer. More terminology is presented and explained, and the experi- 
ments and exercises always relate to the theory presented in Part I of 
the text. The practical understanding of the operation of the synthe- 
sizer gained through studying this part of the text will quickly enable 
anyone to learn how to operate nearly any synthesizer. 

Part III of the text ties this newly mastered instrument into the frame- 
work of traditional musical concepts. The synthesizer is a remarkably 
versatile and flexible tool for the musician, and this part of the book 
explains how the synthesizer can be used to demonstrate and reinforce 
many basic concepts in music and sound. 

The material presented in this book will be best understood if the 
three major parts of this book are studied in the order presented. And 
although there is no substitute for practical experience with a synthe- 
sizer, it is possible to gain a good basic understanding of electronic music 
production by studying the many detailed diagrams in this book. 

The synthesizer represents a major evolutionary step in music and 
music education. It adapts so well to such a wide variety of applications 
that its usefulness is limited only by one's imagination. In this book we 
have offered the tools necessary to begin exploring and enjoying the 
new capabilities of electronic music and synthesizers. 

The authors would like to express their thanks to Dave Fredericks 
of ARP Instruments, Inc., and to Gary Meisner and Dick Peck of Hal 
Leonard Publishing Corporation for their invaluable assistance in- 
writing this book, and to Vivian Hutchins,* editor, and Margaret 
Shepherd for art direction. 

Newton, Massachusetts David Friend 

February 1974 



Introduction: Electronic Sound Synthesis 

Music is an art which uses sound waves as a medium for conveying 
ideas from a musician to a listener. 

A musician uses either the human voice or instruments to create 
musical sounds. The human voice is naturally expressive; the quality 
or timbre of every sound it makes is extremely flexible and capable of 
both subtle and dramatic variations from note to note and within indi- 
vidual notes. The expressive capabilities of the human voice make it 
possible to achieve many artistically useful emotional qualities. 

Music made by musical instruments can also be expressive. However, 
the range of timbres and pitches obtainable from most instruments is 
inherently limited. These limitations must be overcome by the com- 
poser, arranger, and performer to obtain musically pleasing and artistic- 
ally valid qualities. A symphony orchestra, for instance, overcomes 
some of these limitations by using many different instruments, which 
together have a wide range of pitches and timbres. 

Solo instruments, of course, create special problems for the composer 
and performer because of their limited range of sounds. Even complex 
polyphonic instruments like pianos and organs have a relatively limited 
range of timbres and provide very little control over the sound once a 
note has been struck. 

Electronic music synthesizers are the most expressive musical instru- 
ments yet designed, since they allow for the most complete control of 


timbre variations. Like the human voice, it is possible for every note 
played on a synthesizer to have a timbre different from its neighboring 
notes. And it is possible to change the timbre of a note while it is being 
played, either over an enormous range of tone colors, or through very 
subtle changes. And most synthesizers can produce an enormous range 
of pitches, from lower than the lowest organ pipe to higher than the 
human ear can perceive sound. 

The family of synthesizers includes instruments that are capable of 
imitating traditional instruments, and also of creating "new sounds" 
never before obtainable with any instruments. Synthesizers can produce 
pitched sounds, unpitched sounds (wind, thunder, machinery, etc.), and 
sequences of sounds too complex or too fast for performers to play on 
ordinary instruments. 

Two basic ideas are involved in all electronic sound synthesis. The 
first is that acoustical waveforms — virtually any sound you hear — can 
be generated and modified by purely electronic means. In short, any 
sound from the musical sound of a clarinet to the howling of the wind 
outside your home during a thunderstorm can be produced electron- 
ically, given the right kind of equipment. The second idea upon which 
electronic sound synthesis is based is that this sound-generating and 
sound-modifying equipment can be controlled electronically. 

With a synthesizer, you can both invent and imitate sounds. It is 
a scientific mixer, sifter, and producer of sound waves, and it opens 
up truly unlimited scope for experimentation. The "super hand" of 
voltage control carries out your commands instantaneously and with 
great precision; by the time you have finished this book, you will 
know why and how you can achieve the results you want. 

Lesson 1: Waveforms 

Now that we've established the two basic ideas involved in electronic 
sound synthesis, let's back up to the first idea and examine waveforms 
in more detail. Sound is transmitted through several mediums; the 
one we are most concerned with here is air. For example, assume that 
you have an A-440 tuning fork. (Your school music department 
probably has one that you can use.) When you tap this tuning fork 
on the edge of a desk or a table, you are setting up a series of vibra- 
tions that produce the sound you hear. An A-440 fork vibrates at the 
rate of 440 cycles per second, producing the note — or pitch — A above 
middle C. Stated another way, this means that in l/440th of a second, 
the fork makes one complete vibration back and forth (Figure 1.1). 


If you had an exceedingly precise barometer, it would register a 
variation in air pressure during that same 1 /440th of a second 
(Figure 1.2). The displacement of air which the barometer would 
measure is simply a series of air waves set up by the vibration of the 
tuning fork. When these waves reach your eardrum, they set up a 
vibration corresponding to the one created by the natural, or acoustic, 
sound source (Figure 1.3). 

Figure 1.3. Sound waves. 

When someone speaks, sings, or plays an instrument, similar air 
waves are generated, transmitting the sound to your ear. These sound 
waves can be picked up by a microphone and converted into electrical 
signals (Figure 1.4). The electrical signal for each individual sound 
has a particular shape, which can be shown on an oscilloscope. The 
shape of each signal is called its waveshape or waveform. 

The fact that waveforms have a definite shape is an important 
concept to remember. The sound of your voice, the sound of a trum- 
pet, and the sound of an automobile horn all produce waveforms, 
and each of these waveforms has a different shape. Different sounds 
have differently shaped waveforms (Figure 1.5). This is extremely 

important in terms of the sound synthesis you'll be doing before 
long, because the reverse is also true: if the shape of the waveform is 
modified (changed), the sound will also be changed. 

Figure 1.4. Sound-wave transmission. 

1 Cycle 1 Cycle 1 Cycle 

WUU ftrv\ V\AA 


Figure 1.5. Three different waveforms. 

Basic Waveforms 

The most basic classification of waveforms breaks them down into 
two groups: periodic and aperiodic. Periodic waveforms have a repeat- 
ing pattern, as in the case of the waveforms shown for the clarinet, 
the flute, and the brass instrument. Note the marking, "one cycle," 
the following complete cycle, as well as the preceding cycle, look 
very much alike — there is a repeating pattern. The shape of these 
waveforms could, and probably would, change somewhat over a longer 
period of time, but a definite repeating pattern would still be apparent. 

Not all waveforms, however, are periodic. Some waveforms occur 
only once, or at irregular intervals and only upon command. Other 


waveforms are so complicated that your chances of finding any kind of 
repeating pattern are almost nonexistent. In any case, a waveform 
that does not demonstrate any repeating patterns is called aperiodic 
(Figure 1.6). Sounds with aperiodic waveforms have no pitch — for 
instance, those made by thunder, surf, rain, wind, a snare drum, etc. 

Periodic Waveforms 

For now, we'll be most concerned with the periodic type, since most 
musical sounds have a periodic waveform. The shape of four common 
periodic waveforms is shown in Figure 1.7. 

Figure 1.6. 
Aperiodic waveforms. 


A pure, flute-like sound 

A bright, full, brassy sound 

I 1 I I j A bright, but hollow sound 

Jr~ „ T ' much like a clarinet 

n it 

A nasal, "reedy" sound 


Figure 1.7. Periodic waveforms. 

Amplitude and Frequency 

We have established that every sound has a waveform with a par- 
ticular shape. That waveform, however, is not perfectly constant; if it 
were, all that you would hear would be a monotonous tone, with no 
change in the volume, pitch, or timbre (tone quality). We can change 
the volume (amplitude) or pitch (frequency) of a waveform without 
changing its basic shape. 

The amplitude of a waveform is the amount of maximum deviation 
from its "center." In a loudspeaker, this center is the position of 
the loudspeaker cone at rest; in an electrical circuit, it might be the 
condition of zero voltage. The amplitude of the loudspeaker cone's 
vibration would be measured in inches, or fractions of inches, that 
the cone moves back and forth. The amplitude of a fluctuating or 
alternating voltage would be measured in positive and negative volts. 


Thus a voltage waveform that reached a peak of +1 volt and then of 
-1 volt would have an amplitude of 2 volts "peak-to-peak," as shown 
in Figure 1.8. 

The volume control on a record player determines the amplitude 

Amplitude 1 volt Amplitude 4 volts. Amplitude 6 volts Amplitude 8 volts 
peak to peak peak to peak peak to peak peak to peak 

Figure 1.8. Sine waves of different amplitude. 

of the electrical signal delivered to the speaker; the greater the ampli- 
tude of the electrical signal delivered to the speaker, the louder the 
sound from the speaker. For the sake of keeping our terminology 
accurate, then, we might say that the volume control on a record player 
is really an amplitude control since it really controls only the amplitude 
of the electrical signal going into the speaker. It is the amplitude of 
the electrical signal that determines the volume of sound coming out 
of the speaker. 

An important point has now been raised in our discussion of am- 
plitude. You'll save yourself a lot of confusion by thinking of 
amplitude only in connection with electrical waveforms, and of 
volume only in connection with sounds. For example, because there 
are no sounds inside a synthesizer — only voltages — there cannot 
logically be any volume controls. There are, however, a great many 
amplitude controls. In fact, until any electronic device - whether 
it is your stereo or a synthesizer — is connected to a loudspeaker, 
changes in amplitude can be made that will in no way affect the 
volume of the sound you're hearing. More important, in working with 
a synthesizer, changing the amplitude of a waveform in many cases 
will have an audible effect other than a change in volume of whatever 
sound you are creating. Therefore, while a change in the amplitude 
of an audible waveform could change the volume of the sound you're 
hearing, it is more precise and much less confusing to speak of amplitude 
in connection with waveforms, and of volume only in connection with 
an actual change in the loudness of a sound. 


The frequency of a waveform also does not change the basic 
waveshape. We have already established that periodic waveforms 
have a repeating pattern or cycle. The rate at which that cycle is 
repeated is the frequency of the waveform. For example, when 
discussing the tuning fork we said that it was tuned to A-440, pro- 
ducing A above middle C. This means that the waveform frequency 
required to produce this note is 440 cycles per second. If the fre- 
quency were lower, the pitch of the note you hear would be lower. 
If the frequency were higher, the pitch would be higher. There is an 
international standard unit of measurement for frequency: this unit 
is "Hertz" (abbreviated as Hz) and it is defined as one cycle per 
second. Thus, the A pitch that we have been discussing is produced 
by a frequency of 440 Hz. The prefix "Kilo-" means "one-thousand," 
and a frequency of 1,000 cycles per second is called "one KiloHertz," 
or 1 KHz. 

We've established that changing the frequency of the waveform 
changes the pitch. However, just as with amplitude and volume, it 
is important to keep the ideas of frequency and pitch separate in 
your mind. Frequency is a characteristic of all vibrations, whether 
mechanical or electrical, while pitch is a characteristic of the way 
human beings perceive acoustic vibrations between approximately 
20 Hz and 20 KHz. Another way to say this is that every pitch (a 
sound we can hear associated with a particular frequency) is produced 
by some frequency, but many frequencies produce no pitch. This is 
because there are many frequencies above and below the range of 

Figure 1.9 

human hearing. If you were able to try the experiment with the 
tuning fork, by tapping it on the side of a desk and listening to the 
pitch produced, you already know that you can hear the frequency 
440 Hz. On the other hand, you would not be able to hear 4.40 Hz — 
it is far below audibility. Nevertheless, subsonic, or low, frequencies 
are very useful in synthesizing many sounds, and it is for this reason 
that frequency must be thought of as being related to, but distinct 
from, pitch. 


One final point concerning amplitude and frequency: changing the 
amplitude or frequency of a waveform does not alter the basic shape 
of the waveform. In Figure 1.10, illustrating the three sawtooth 
waves of different frequencies, you'll notice that the basic shape of 
the waveform does not change at all - it is simply narrower and 
appears more often in the same period of time as the frequency 
rises. The essential characteristics of the shape are still present. The 
same is true of amplitude; if you examine the three sine waves of 
different amplitudes, you'll see that the general shape of the wave- 
form does not change. It still looks like a sweeping S-curve, with 
just slightly different dimensions. Changing the frequency or ampli- 
tude of a wave will not count as a change in its shape (Figure 1.11). 

*■ .1 second *~ Figure 1.10. 

Period ~ .050 Frequency = ^ Q = 20 Hz Sawtooth wave of 


three different frequencies. 

m .1 second * 

Period = .025 Frequency = q~ ^ = 40 Hz 


m .1 second * ^ 

Period = .0125 Frequency = = 80 Hz 



Figure 1.11. Sine waves of three different amplitudes. 


Changes in the shape of an audio waveform are generally associated 
with changes in the tone quality, or timbre, of the sound produced. 
Generally speaking, timbre is the subjective quality of a tone which 


enables the listener to distinguish between it and other tones which 
may have the same pitch and/or loudness. For example, trumpets and 
clarinets produce sounds of different timbres but can be equally loud. 
Since human perception has limits, changes in the shape of a wave- 
form are possible that might not be perceived by even the most 
practiced ear. However, any time that a change in timbre is perceived, 
you can be sure that a corresponding change must be reflected in 
the shape of the waveform. 

Summary of Terms 

Based upon what we now know, we can set up the following chart 
summarizing these approximate relationships: 

Perceived subjective changes in: Correspond to physical changes in: 
Volume Amplitude 
Pitch Frequency 
Timbre Waveform or Waveshape 

This is worth remembering because, as emphasized earlier, synthe- 
sizers work with the qualities listed in the right-hand column, whereas 
we perceive the results as changes in those subjective qualities listed 
in the left-hand column. With practice, you will learn to translate 
easily from the language of volume, pitch, and timbre to the language 
of amplitude, frequency, and waveshape. 


1. If you use a tape recorder to record a clarinet playing an A-440 Hz 
and then slow the tape down to half speed, what would the pitch be 
and would the waveform shape change? Why? 

2. If you talk with your face in a pillow, does the pillow alter the 
waveshape of your normal speaking voice? Why do you think so? 

3. Are the following sounds periodic or aperiodic waveforms: 

(a) a car horn; 

(b) striking a garbage-can lid; 

(c) rustling a newspaper; 

(d) a squeaky door hinge? 

4. What is the difference between volume and amplitude? 

5. If a waveform repeats itself once every 1/100 sec, what is the 
frequency of the waveform? Is this a low-pitched sound or a high- 
pitched sound? 

6. What happens when a sound gets so low in pitch that you can 
no longer hear it? What does it sound like then? How low do you 
think you can hear pitch? 

Lesson 2: Overtones and Harmonics 
The Fundamental 

Given the information that has now been presented, we can ask: 
"Why do differently shaped waveforms produce different sounds?" 
The principal reason is because these differently shaped waveforms 
have different overtones. Overtones can best be defined by example 
and illustration. Assume that you are listening to a particular pitch — 
low A, for example, at 110 Hz. The acoustic effect produced by a 
single vibration of this frequency would be described as a pure 
sound — one with no harmonics, or overtones. A sine wave is a pure 
sound. However, most natural sounds, such as the vibrating string of 
a violin, do not consist of just a single, simple vibration. Virtually 
all musical instruments produce" complex, composite sounds, consisting 
of the main sound — the fundamental — plus a number of additional 
pure sounds of lesser amplitude called overtones. 
Harmonic Frequencies 

In most musical instruments, the overtones consist of vibrations that 
are simply multiples of the fundamental frequency. For instance, an A 
(110 Hz) would have overtones at 220 Hz, 330 Hz, 440 Hz, etc. Over- 
tones like these, that are simple multiples of the fundamental, are called 
harmonics. It is the relationship of the relative amplitudes of the differ- 
ent harmonics that allows us to distinguish between two sounds with 
the same pitch. For instance, one of the reasons a tuba sounds different 
from a bassoon is that the harmonics produced by the two instruments 
are different in amplitudes with respect to the fundamental. Strings, 
trumpets, and reeds produce tones which are relatively rich in harmon- 
ics, while flutes and French horns produce sounds with relatively few 
harmonics. Figure 2.1 shows these harmonics with their pitches. 


Fundamental 110 Hz Low A 
Sine Wave 

2nd Harmonic 220 Hz A, one octave 

above 110 Hz 

3rd Harmonic 330 Hz E above middle C 

4th Harmonic 440 Hz A above middle C 

^\J\f\J\f^ 5th Harmonic 550 Hz C sharp above A -440 
Figure 2.1. Harmonic frequencies. 


Therefore, the various regular waveforms differ from each other by 
their harmonic content and this difference accounts for their differ- 
ent sounds. Figure 2.2 illustrates the harmonic content of the sine, 
square, sawtooth, and pulse waves. 






— i — 0 — 



m — 










of the 

1 1 , 




F. 3rd 5th 7 th 9th 

F.234 56 7 89 

F. 23456789 

Figure 2.2. Harmonic series. 

You'll note that the square wave is made up of the fundamental 
frequency plus all odd-numbered harmonics, while the sawtooth and 
narrow pulse are made up of the fundamental plus all harmonics. 
The essential difference between the sawtooth and the narrow pulse 
waves is that as you listen to the two, you will notice that the pulse 
wave sounds brighter. This is because the amplitude of the pulse-wave 
harmonics is greater than the amplitude of the harmonics of the 
sawtooth wave. Consequently, the greater amplitude of the higher 
harmonics results in the narrow pulse wave having the brighter 
sound to your ear. 
Other Overtones 

Some instruments, especially instruments involving the striking of 
metal, like gongs, chimes, triangles, and bells, produce overtones that 
are not exact multiples of the fundamental. Sometimes, as with a 
Chinese gong, there are so many overtones that it is virtually impossible 
to tell which is really the fundamental pitch. The frequencies of these 
overtones can be in very odd relationship to the frequency of the 
fundamental, like 7/2 or 9/4. 


Since the vibrations created by the oscillators (tone generators) on a 
synthesizer have only simple harmonics, a device called a "ring 
modulator" is provided to create overtones which are not harmonics, 
thereby permitting the synthesis of metallic sounds. 


As you might expect, the chart showing the four waveshapes — 
sine, square, sawtooth, and pulse — does not represent every con- 
ceivable waveshape. As Figure 2.3 shows, there are a number of inter- 
mediate waveshapes that consist of the fundamental plus fewer harmon- 
ics than are found in the square, sawtooth, and narrow pulse waves. 

Fundamental, plus 2nd Harmonic Fundamental, plus 3rd Harmonic 

Fundamental, plus 2nd, 3rd, Fundamental, plus 3rd and 5th 

4th, and 5th Harmonic Harmonic 

Figure 2.3. Waveshape created by adding harmonics. 

It is useful to note that as more higher harmonics are added, the 
corners on the corresponding waveforms become sharper and more 
abrupt. Thus, looking at a waveform that has smooth, rounded fea- 
tures should tell you that this waveform will sound mellow or 
flutelike (Figure 2.4). On the other hand, a jagged waveform with 
sharp corners will always produce a bright, buzzing sound. Figure 2.5). 

Figure 2.4. Mellow sound has Figure 2.5. Bright sound has 

smooth-looking waveform. jagged, angular waveform. 

Based upon what you've now read concerning harmonics, you have 
the basic knowledge necessary to understand the significance of the 
following point: the sine wave is the simplest waveform, and all 
other waves are actually combinations of sine waves of related fre- 
quencies, called harmonics. Stated another way, complex waveforms 
can be synthesized by adding together a number of simple waves; 
logically enough, this process is called additive synthesis. Remember 
this term, because the converse of this concept will appear in the 
following discussion of filtering. 



1. What is the relationship between timbre and harmonics? 

2. If you change the harmonics in a sound, will the timbre change? 
Will the waveform change? 

3. Which waveform would sound brighter: a waveform with smooth, 
flowing features or a waveform with sharp, jagged corners? 

4. Which instrument produces more harmonics: a slide whistle or 
a kazoo? 

5. What is meant by "additive synthesis" ? 

Lesson 3: Filters and Filtering 

Subtractive Synthesis 

Just as complex waveforms may be created by adding a number of 
simple waves together, it is possible to simplify a complex wave by 
filtering out certain frequencies. This process is called subtractive 
synthesis. This concept is extremely important, as it will form the 
basis for much of the sound synthesis you will be doing on the 
ARP Odyssey and, in fact, with any synthesizer. 

The tone controls of a stereo are no more than simple filters 
capable of removing or attenuating the high (treble control) or low 
(bass control) frequencies. Synthesizers, on the other hand, may em- 
ploy as many as four types of somewhat more sophisticated filters. 
These filters, and their specific functions, are shown below. 

An ideal low-pass filter would pass all frequencies up to the cutoff 
point and then completely eliminate all frequencies above this cutoff 
point. A graph showing how an ideal low-pass filter works is shown in 
Figure 3.1 (a). In practice, however, it is impossible to create an ideal 
low-pass filter. Instead, low-pass filters (such as the one on the Odyssey 
synthesizer) pass all frequencies up to the cutoff point, and then 
gradually reject the frequencies above this cutoff point. This gradual 
attenuation of the higher frequencies is called "rolloff." The response 
of a low-pass filter with rolloff is shown in Figure 3.1 (b). 

Some low-pass filters, like the one in the ARP Odyssey, have an 
additional characteristic called resonance. Adding resonance to a low- 
pass filter causes the filter to emphasize a band of frequencies just at 
the cutoff point. The response of a low-pass filter with resonance is 
shown in Figure 3.1 (c). Resonance is extremely important in synthe- 
sizer filters. Almost every mechanical instrument has its own character- 
istic resonances, and in order to simulate most natural sounds, the 
synthesizer must be able to duplicate these resonances. 


3.1 (a) Idealized LPF 3.1 (b) LPF with Rolloff 3.1 (c) LPF with Resonance 

Low-pass filters with resonance occur commonly in nature. When 
you talk through a long pipe, for instance, you will notice that your 
voice is muffled because the pipe is a low-pass filter, but your voice also 
has a kind of nasal quality which is characteristic of resonance. A pillow 
would be an example of a low-pass filter without resonance. Your voice 
is simply muffled by the pillow. 

Low-Pass Filter. Each filter name is descriptive. As this name would 
imply, the low-pass filter passes all frequencies below a certain cutoff 

with Rolloff 



of signal ^ % You hear these 
passed frequencies 

filter . 1% \ Fig 1116 3 - L 

20 Hz 20 KHz Low-pass filter function. 


High-Pass Filter. This filter passes those frequencies above the 
filter cutoff frequency (Figure 3.2). 




You hear these 

1% I Figure 3.2. 

20 Hz 20 KHz High-pass filter function. 


Band-Pass Filter. The band-pass filter passes only a certain band 
(range) of frequencies, eliminating or attenuating those frequencies 
both above and below the band being passed (Figure 3.3). 


Figure 3.3. 

20 Hz 20 KHz Band-pass filter function. 

Band-Reject Filter. This filter passes all frequencies except a certain 
band of frequencies. Thus, the function that this filter performs is 
exactly opposite to that of the band-pass filter (Figure 3.4). 

100% f 

10% You hear these \ / You hear these 

frequencies \ | frequencies 

1% 1 f Figure 3.4. 

20 Hz 20 KHz Band-reject filter function. 

The most useful filters are the low-pass and the band-pass filters. 
This is because the bodies of most natural, or acoustic, instruments 
(the tubing and bell of a trumpet, the wooden box of a violin) are 
low-pass and band-pass filters. The ARP Odyssey has one voltage- 
controlled filter which can be used as either a low-pass or a band-pass 
filter. In addition, it has a separate manually controlled high-pass 

Waveform Modification 

The important point to remember here is that the filter (s) of a 
synthesizer are waveform modifiers, capable of changing the shape, and 
therefore the sound, of the waveform passing through the filter. 
Basically, this is done by filtering out certain components of the 
waveform. For example, if all the harmonics of a square, or sawtooth, 
or pulse wave are filtered out, the result will be a sine wave. Therefore, 
by filtering — or subtractive synthesis — the shape of the waveform 


has been reduced to its fundamental component. Passing a square 
wave through an idealized low-pass filter will change the waveshape. 
In Figure 3.5 (a) through (d), the cutoff point of the filter is made 
lower and lower, cutting off more of the harmonics (Figure 3.5). 
This selective removal of the overtones or harmonics will become 
extremely important as you begin synthesizing both instrumental 
and electronic sounds. 












Figure 3.5' Filtering out harmonics with an Ideal Low-pass Filter 

Notice that the same waveforms can be created by filtering the 
harmonics out of a square wave in Figure 3.5 as were created by adding 
together sine waves in Figure 2.3. 

Because the low-pass filters on synthesizers only approximate an 
ideal filter, we cannot produce the exact waveforms shown in Figure 
3.5. To your ear, however, the approximation is very close indeed. 
The basic principle of subtractive synthesis— shaping a waveform by 
filtering— will allow us to create an enormous range of musical sounds 
with the filters available on synthesizers such as the ARP Odyssey. 


L When you talk through a pillow, your voice becomes muffled. 
Do you think the pillow is a filter? What kind? Why? 

2. Would passing a waveform through a high-pass filter make the 
sound brighter or muffled? 

3. When you listen to music over a telephone, it doesn't seem to 
have good bass or good treble. What kind of filter is the telephone? 

4. Describe the process of subtractive synthesis. 


Lesson 4: Low-Frequency Waveforms 

Subsonic Frequencies 

Up to this point we have been largely concerned with periodic, 
audio waveforms. A synthesizer, however, can generate and modify 
both audio frequencies and subsonic, or low, frequencies. You'll 
remember that we call those frequencies between approximately 
20 Hz and 20 KHz "audio frequencies." This is because it is only 
when physical vibrations are within that range that we perceive them 
as sounds having a definite pitch. However, it is possible to hear 
things happening at frequencies lower than 20 Hz — but you'll hear 
them as separate and repeating sounds, not as continuing tones or 
noises. For example, you can plainly hear your heartbeat by listening 
through a stethoscope. Although it would be rather slow for a pulse 
rate, let's say that your heart is beating 60 times per minute. That's 
the same as saying it beats once every second, or at a frequency of 
1 Hz. Now imagine your heartbeat gradually increasing to 120 per 
minute, or 2 Hz. (Your heart frequently beats this fast during mod- 
erately strenuous exercise, but 2 Hz is still well below the limits 
of audibility.) However, assume that your pulse rate could run on 
up to 1,200 heartbeats per minute. At 1,200 per minute, your heart 
would be beating 20 times per second, or at a frequency of 20 Hz. 
At this point, or at somewhere just above 20 Hz, you would lose 
your sense of hearing individual heartbeats, and instead would 
begin to hear a very low pitch, which would gradually rise with the 
rising frequency of your heartbeat. As your heartbeat slowed, the 
opposite would happen: first, you would hear a falling pitch, then a 
gradual transition to a state of hearing no pitch at all, but rather 
separate and countable beats, or events. 

We can use another example: playing a note on a piano, at a rate 
of one note every second, produces a series of low-frequency events 
at 1 Hz. These events do have a certain pitch because each event is 
the occurence of an audio-frequency vibration. A graph of everything 
that is happening as you play three successive notes would look like 
Figure 4.1: 

tfFast vibrations of the string 

Figure 4.1. Low-frequency events. 


The low-frequency waveform involved would look like Figure 4.2: 
^. Decay of sound 

Figure 4.2. Low-frequency waveform. 

And the audio frequency that produces the pitch you hear would 
look like Figure 4.3. 

Figure 4.3. Audible part of sound in Figure 4.1. 

Note that when an event, or series of events, is represented by a 
graph, you can derive the shape of the low-frequency waveform by 
simply connecting the highest points on the audio-, or higher, fre- 
quency waveform. Now we can see that events have shapes, and the 
shape of any event is the shape of the low-frequency waveform that 
can "produce" that event. For example, playing a staccato tune on 
an organ, where decay is almost instantaneous, would produce a 
series of different pitches that would look something like Figure 4.4: 

Figure 4.4. Staccato organ notes. 

And a series of events would be produced having the shape in 
Figure 4.5. 

i i i i i i 

Figure 4.5. Event shape of Figure 4.4. 

Playing the identical tune on a guitar would produce the same 
pitches (Figure 4.6): 

Figure 4.6. Staccato notes on guitar. 

But once a guitar's strong vibration trails off, the events would 
have an altogether different shape (Figure 4.7). 

Figure 4.7. Event shape of Figure 4.6. 



The shape of an event is called its envelope or contour. In the 
example given above, the notes played on the organ have a different 
envelope than the same notes played on the guitar. Figure 4.8 shows 
some other possible envelopes: 






(This can be 
done easily 
on a 

Figure 4.8. Six possible envelopes. 

Any low-frequency waveform may be used to produce events and 
to give them a shape. Usually, however, a synthesizer will have one 
or more devices designed specifically to generate low-frequency 
waveforms suitable for giving events a shape. These devices are called 
envelope generators. The output of an envelope generator is aperiodic; 
instead of appearing over and over again, as a low-frequency periodic 
waveform would, the low-frequency waveform from an envelope 
generator appears only when the envelope generator is triggered (for 
instance, by striking a note on the keyboard). 


1. Honking a car horn would produce an event shaped like Figure 
4.5. Draw the shapes or envelopes of the following: a piano note (first 
a high note, then a low note), a tuba, a drum roll, a harp, a cymbal 
crash, and surf. 

2. If you use a tape recorder to record a clock which ticks once 
every second and then speed up the tape 100 times faster, what 
would you hear? 

3. If you record a bassoon playing low A-110 Hz, and then slow 
the tape recorder down 100 times, what would you hear? 


Lesson 5: Voltage Control 

You will recall that at the beginning of Part I, two ideas were 
presented: the first, that acoustical waveforms can be generated and 
modified by purely electronic means. Based on what you've read here 
concerning waveforms, amplitude and frequency, timbre, harmonics, 
and filtering, you are beginning to have a basic understanding of how 
this is possible. However, the second idea presented at the beginning 
of this section — that sound-generating and sound-modifying equip- 
ment can be controlled electronically — has been discussed only in 
indirect ways. With the discussion of the envelope generator, we have 
an excellent example of voltage control which now can be discussed 
in more detail. 
Envelope Generators 

An envelope generator is generally used to create a signal which can 
control the Voltage Controlled Filter or the Voltage Controlled Ampli- 
fier on a synthesizer, thereby giving the sound produced by the synthe- 
sizer the desired shape or "envelope." Let us consider the most common 
use of the envelope generator— controlling the filter. The voltage con- 
trolled filter (abbreviated VCF) on a synthesizer can, of course, be 
opened and closed manually. (The terms "open" and "closed" refer to 
raising or lowering the cutoff point of the filter to let more or fewer 
harmonics pass through).. However, in order to synthesize most natural 
instrumental sounds— or even more broadly, most musical sounds— a very 
sophisticated, programmable way of automatically opening and closing 
the filter is required. This is necessary because the harmonic com- 
ponents of most instrumental tones are not constant from the begin- 
ning of the sound to the end of the sound. For example, as you blow 
into any wind instrument, a certain amount of time is required simply 
to start the tone. This is called attack time. Consider the relatively 
slow attack time of a tuba for example. This would contrast with the 
fairly fast attack of a guitar, which sounds a tone almost instanta- 
neously when the strings are struck by the guitarist's pick. During 
the attack of a tuba, for instance, the sound builds gradually during 
about 1/5 second after you start blowing. Not only does the sound 
get louder during this period, but it also gets richer in harmonics and 
consequently brighter in sound. During the attack of a guitar, all the 
harmonics seem to sound from the instant the string is plucked, and 
from that point on they decay, the highest harmonics disappearing 

The envelope generator is hooked up to the filter in a synthesizer 
so that the attack and decay characteristics of any sound can be 
programmed on the envelope generator's controls, and the filter will 


then be "opened" or "closed" automatically by the envelope gene- 
rator. To synthesize a tuba sound, for instance, we would want to set 
the controls on the envelope generator so that the sound would build 
up during the first 1/5 second. When the envelope generator is trig- 
gered by pressing a note on the keyboard, the envelope generator 
produces a voltage which automatically opens the filter during the 
first 1/5 second. The result is that as the filter opens, it lets more 
harmonics pass through and creates an attack which sounds very 
much like the attack on a tuba. You can also simulate the faster 
attack time of the guitar, bringing all of the harmonics that create 
the distinctive timbre of the guitar almost immediately. This is done 
by simply decreasing the attack time so that the filter opens more 
quickly, thus permitting all of the necessary harmonics to sound 
very quickly as each note is played. 

The point here is that the low-frequency, aperiodic waveform that 
the envelope generator provides is a voltage. This voltage opens the 
filter for you, in a more precise manner than any type of manual 
control could ever provide. By adjusting the attack, decay, sustain, 
and release times, this voltage will open, hold open, and then close 
the filter in a manner characteristic of the sound you are synthesizing. 
This voltage control of the filter provides programmable control 
over the timbre of the waveform from the instant that the tone 
begins to sound to the final moment before it dies out. 

All synthesizers must have envelope generators. There are different 
types of envelope generators. The Odyssey, for instance, has two 
envelope generators. One of these has four controls, for Attack, Decay, 
Sustain, and Release, and is called an ADSR type of envelope generator. 
(See Figure 5.1). The second envelope generator on the ARP Odyssey 
has only two controls, Attack and Release. This simpler type of envelope 
generator is called an AR envelope generator. (See Figure 5.2). 

Attack Time 

Figure 5.1. Four adjustable parameters Figure 5.2. Two adjustable parameters 
on an ADSR envelope generator on an AR envelope generator 


Other Voltage Controllers 

The envelope generator is only one example of voltage control as 
it is employed in synthesizers. You'll notice that each of the oscilla- 
tors (the waveform generators, or sound sources, of the Odyssey), 
the filter, and the amplifier of the Odyssey are marked "voltage- 
controlled." The keyboard is also a voltage controller, providing a 
voltage that is applied to the oscillator (s) to determine the frequency 
of the pitch that you'll hear. Certainly you could move the frequency- 
control slider of the oscillator manually to change its pitch — but 
could you move it quickly and accurately enough to play even a 
moderately difficult piece of music at a reasonable tempo, or speed? 
The answer, of course, is no. Without the voltage control that the 
keyboard provides, it would be difficult to play even the simplest 
tune without having some alternative to moving the frequency- 
control slider up and down by hand. Voltage control, as you'll soon 
determine for yourself, is an extremely important extension of your 
ability to control the various functions of the synthesizer. 

The whole point of making voltage-controlled circuits on a syn- 
thesizer is so that the synthesizer's electronic circuits can "talk" to 
each other in a language they understand and can respond to — 
voltages. A human being can turn a knob or flick a switch, but an 
electronic circuit must use voltages to control other electronic circuits. 
For instance, let's say that you are listening to a tone generated by 
an oscillator. If you want to change the pitch of that tone, you can 
physically move the control labeled "frequency." But if you want 
to change the pitch of the oscillator by playing on the keyboard, 
how does the keyboard tell the oscillator to change its pitch? The 
answer, of course, is that the keyboard produces a voltage which is 
fed into the oscillator and tells the oscillator to change its pitch. 
Because the oscillator is designed to react to control voltages from 
other circuits (like the keyboard), the oscillator is said to be a 
"Voltage-Controlled Oscillator." 

From our discussion, an important concept has emerged — a 
concept so basic to an understanding of voltage control that it de- 
serves emphasis: in synthesizing almost any sound you will be em- 
ploying two types of voltages: (1) an audio signal, and (2) a control 
voltage. The audio signal provides the basis for the sound that you'll 
hear. This signal may be in the form of a sawtooth wave, or a sine 
wave, or a square wave, or a pulse wave. Whatever the waveshape, 
this is the basic raw material for the sound that you will hear. In 
short, it is an audio frequency — between approximately 20 Hz and 
20 KHz. 


Control Voltages 

Unless you are creating an extremely simple effect, however, you 
will also be using control voltages. While you don't normally "hear" 
the control voltage, you may hear its effect upon the audio signal. 
For example, pressing down one key on the keyboard of the synthe- 
sizer and then pressing down a different key will cause the audio 
frequency the oscillator is producing to change, thus changing the 
pitch of the sound you're hearing. While you did not "hear" the 
control voltage that was sent to the oscillator to cause this change, 
you did hear its effect. Similarly, you may hear a particular instru- 
mental sound that has a pleasant vibrato (a gentle, pulsating effect). 
In synthesizing this sound, you will actually hear the audio frequency 
produced by the oscillator; you will not hear the low-frequency sine 
wave creating the vibrato (this is the control voltage) but you will 
hear its effect. This concept will become increasingly clear as you 
begin using the voltage-controlled functions of the synthesizer pre- 
sented in Part II of this text. 

Frequency Modulation and Amplitude Modulation 

One final basic concept must be presented before moving on to 
Part II and the actual operation of the ARP Odyssey. This concept 
is called "modulation." To modulate a waveform is to change it 
periodically, following the pattern of another, waveform. If the 
change in the first waveform is a change in frequency, it is called 
frequency modulation; if the change is in amplitude, it is called 
amplitude modulation. 

Here is a simple example of frequency modulation. Suppose we 
begin with a simple oscillation — a sine wave at the audible fre- 
quency of 100 Hz. Now suppose that we wanted the frequency (the 
pitch that you are hearing) of that sine wave to rise gradually to 
200 Hz, then fall gradually to 100 Hz and begin this cycle again. To 
do this, we would frequency-modulate the 100 Hz square wave with 

Figure 5.3. Frequency modulation. 





a low-frequency sine wave — say, at the rate of 1 Hz. As a result, you 
would hear the pitch of the sine wave rise gradually from 100 Hz 
to 200 Hz once every second. Each time it reached 200 Hz, it would 
begin to fall to 100 Hz and repeat the cycle. A graph of the changes 
in frequency is shown in the Figure 5.3. 

Amplitude modulation, on the other hand, would produce something 
like Figure 5.4. 



Figure 5.4. Amplitude modulation. 

In each illustration you will see that what is happening does conform 
to the definition established earlier: the first waveform (the sine 
wave) is changing systematically following the pattern of another 
waveform (the 1 Hz sine wave). 


1. If an envelope generator can be hooked up to a voltage-controlled 
filter to produce an attack and decay like a tuba but automatically 
opening and closing the filter, what would happen if that same envelope 
generator were hooked up to a voltage-controlled oscillator that was 
producing a pitch? 

2. Explain the concept of "voltage control." How can voltage con- 
trol be more effective than manual control? 

3. If you play a trill on a clarinet, are you modulating the frequency 
or the amplitude of the waveform? What is the shape of the modulat- 
ing signal? 

4. If you sing a note while patting your mouth with your hand, 
what kind of modulation do you get? 

5. A modern electronic police siren is actually a voltage-controlled 
oscillator hooked up to a loudspeaker. A siren is an example of what 
kind of modulation? What do you think the control signal used to 
create this modulation looks like? 


Section 1: Basic Operational Features 

Having now completed Part I, you should have acquired a basic 
understanding of generating and modifying sounds by purely elec- 
tronic means. This section is designed to guide you in applying this 
knowledge in a practical manner through the actual operation of the 
ARP Odyssey synthesizer. 

Block Diagrams and Patches 

To begin, study the diagram shown in Figure 1.1.1. The control 
panel of the Odyssey is organized much like an assembly line. Gener- 
ally speaking, the sounds you will create will be assembled beginning 

Figure 1.1.1. Control panel organization. 


with the raw material — the oscillators and/or noise generator — loca- 
ted on the left side of the panel. Your "product" will then move from 
the left to the right side of the instrument, passing through the various 
stages of the assembly line that are required to shape the raw material 
into the finished sound you are seeking. You will observe this left-to- 
right movement as you begin experimenting with the Odyssey. 
Similarly, the method used to record visually a signal's path across 
the instrument reads left-to-right. These visual representations of the 
assembly line are called block diagrams. Basic block diagrams for a 
trumpet sound, as produced by a trumpet and by the Odyssey, are 
shown in Figure 1.1.2. Notice that the audible signal (waveform) path 


of the lips 
on mouthpiece 

and bell 


of horn 



(Shapes attack 
and decay) 





(Controls pitch) 





(Shapes attack and decay) 

Figure 1.1.2. Block diagrams for conventional trumpet and Odyssey trumpet sound. 


runs horizontally, left to right. The control paths run vertically — as, 
for example, in the case of the keyboard controlling the pitch of the 
Odyssey's oscillator (s), or in the use of the envelope generator to 
control the shape of the sound. 

Within the Odyssey, of course, the gathering of the raw materials 
and forming of an assembly line is done electronically. The inter- 
connection of the various functions is known as creating a patch. The 
block diagram shown in Figure 1.2 is a trumpet patch. In a similar 
manner, it would be possible to diagram any other patch that you 
might create. It is important that you begin thinking in terms of this 
sequential left-to-right organization of the Odyssey's controls. Not 
only will it facilitate working with the synthesizer, but it will also 
help you in remembering the essential elements of the patches you 
will be creating. You'll find this useful in freeing yourself from having 
to record the control positions for every sound you want to remember. 
Instead, seek to understand what functions are being employed — the 
oscillator(s), the filter(s), the various controllers — so that you can 
reassemble the sound conceptually rather than by rote. 

The Control Panel 

In Figure 1.1.3 you can see that the diagramming on the control 
panel of the Odyssey is designed to assist you in mentally assembling 
the patches you'll be creating. As in the case of the block diagrams 
shown in Figure 1.1.2 audio signal paths run horizontally, left to right. 
Control paths run vertically, entering the bottom side of the boxes 
which label the functions of the Odyssey. 

Generally speaking, each function on the Odyssey is set aside in an 
area of its own, separated from the other sections of the instrument 
by a line running from the top to bottom of the control panel. We'll 
call these areas "panels," so that when we speak of the "second panel 
from the left" you'll understand that the general area being examined 
or discussed is the Voltage Controlled Oscillator No. 1 panel. 


Now let us examine the basic controls of the Odyssey more closely. 
Note the row of switches along the bottom of the control panel. These, 
and similar switches elsewhere on the panel, are two-position switches, 
like an off/on light switch. Some of these actually are off/on switches, 
while others (the majority of them) are selector switches. The selector 
switches let you choose between two interconnections, depending 
upon which position the switch is in.Figurel.l.4shows the first selec- 
tor switch on Panel 5. 



— |scdss 
— ||azE]i||s| 

HI mil 

— Ilm^ 









"UP" position 
selects the 
Noise Generator 

"DOWN" position 
selects the 
Ring Modulator 

Figure 1.1.4. Selector switch. 

Slide Controls and Attenuators 

The Odyssey also has a number of slider-type controls. These con- 
trols may be divided into two groups: (1) slide controls, which govern 
some function, as in the case of the VCO frequency controls, and 
(2) attenuators, which act much like faucets, controlling the amount 
of signal that is allowed to flow along the particular signal or control 
path you are using (Figure 1.1.5). 

The controls on the Odyssey are color coded to associate them with 
the appropriate circuits that they control or are connected to. The 
color code, by function, follows: 













All others 


The controls which function as input attenuators (that includes all 
the controls along the bottom row except the two pulse-width and 
S/H lag controls on the VCOs and the ADSR) are color coded accord- 
ing to the function connected to each slider when the selector switch 
beneath the slider is in the "up" position. 



You'll find that most slide controls along the bottom row are 
attenuators except for the ADSR, S/H lag, and pulse-width controls. 
All of the controls along the upper row are simply slide controls gov- 
ering a particular function. 

As you begin to experiment with the switches and sliders in con- 
necting the various components of the Odyssey, you'll quickly realize 
that the number of patches is actually infinite. Each full patch, how- 
ever, requires that you employ at least three functions of the Odyssey: 
(1) a signal source, (2) a signal modifier, and (3) a controller. The 
block diagram in Figure 1.1.6 again illustrates the trumpet/Odyssey 
comparison shown earlier (Figure 1.1.2). Note the signal source, 
modifier, and controllers. The following material will deal with each 
of these three components — sources, modifiers, and controllers — in 
detail, allowing you ample opportunity to explore on your own. 


Figure 1.1.6. Simple block diagrams. 

of the lips 
on mouthpiece 


& bell of horn 


(Control Pitch) 


Trumpet sound 

















Trumpet sound 


Section 2: Signal Sources 

On the Odyssey, audio signals in their raw form are generated by 
the noise generator and the two voltage-controlled oscillators — VCO-1 
and VCO-2. The raw signals produced by these signal sources are later 
processed by other circuits to produce a controllable audio output. 
Let's examine one of these signal sources now. 

Lesson 1: Noise Generator 

Using Figure 2.1.1 as a model, set all the controls on your Odyssey 
as shown. Notice that all of the two-position switches are in the "up" 
position. The general setup in Figure 2.1.1 will give you a basic patch 
from which to work. Each time an experiment is completed, return 
all the controls to this position and you'll be ready to begin again. 

Now, turn on the Odyssey (red switch in the upper right-hand 
corner) and locate the Noise Generator switch in the upper part of the 
panel to the extreme left (see Figure 2.1.2). Check to make sure the 
switch is in the "up" position, labeled "White." The first attenuator 
we will use is the white one labeled "Noise," located in the Audio 
Mixer section (the second panel from the right in Figure 2.1.2). Re- 
member, the attenuators are like faucets, setting the amount of a signal 
that is allowed to pass through. As you push the attenuator higher, 
the sound becomes stronger. Try it. Experiment by moving the atten- 
uator up and down through its entire range of travel. 

Noise can be used to synthesize many of the everyday sounds we 
hear — surf, thunder, rain, motors, and even wind. For a moment, 
think about the characteristics of the sound of an ocean wave — how 
it gradually begins, building to a climax with a culminating crash of 
water, then falling back and fading. Let's attempt to reproduce this 

Experiment 1: Noise 

Set the Noise Generator switch on "White" and slowly raise the 
white attenuator, gradually making the "wave" stronger. About half- 
way up, move the attenuator rapidly to the top, and then bring it 
back down at a steady rate. Did you create a surf effect? Figure 2.1.4 
provides a graphic description of what you accomplished. 

White Noise and Pink Noise 

You may have asked yourself, "What kind of waveform does the 
Noise Generator produce?" To answer this question, let's return to 
the earlier discussion of waveforms, in which they were classified under 





1 °- IT 


o jt 
5f j 


8 S3 

o a. 

* 5 3 




i % 


si3 a 






o S 

| Si 


Wave Figure 2.1.4. Listening to white noise, 
crashes ... , 

two general headings — periodic and aperiodic. The Noise Generator 
produces an aperiodic waveform which is completely random. You 
can think of white noise as a waveform in which the chances of finding 
any particular frequency are equal to the chances of finding any other 
frequency. This sort of waveform is called white noise by analogy to 
white light, because it contains all frequencies just as white light con- 
tains all colors. 

Experiment 2: Filtering Noise 

Return to the Odyssey and move the Noise Generator switch from 
White to Pink. Listen to the sound carefully. How does it differ from 
white noise? As you may have guessed, pink noise is created by filter- 
ing out some of the high frequencies of white noise, thereby allowing 
the lower frequencies to become more predominant. Human hearing 
tends to give undue prominence to the higher frequencies in the white 
noise signal, so that it sounds like steam escaping from a radiator. 
By filtering white noise, we achieve an effect (pink noise) that has a 
frequency content in which we hear all frequencies, the lows as well 
as the highs. Pink noise can be said to have equal energy per octave. 
Don't hesitate to experiment with both white and pink noise. 


1. If there were such a thing as "red" noise, what do you think 
it would sound like? 

2. Name as many sounds as you can that are based on noise. Would 
you be better off starting with white noise or pink noise to make each 
of these sounds? 



»1 tCz: 


* | 54 ' ' | 4: \ J 



Lesson 2: Audio Oscillators 

Moving from the noise generator with its aperiodic waveforms, let's 
now examine two signal sources which generate periodic waveforms. 
These signal generators are called oscillators. Your Odyssey includes 
two oscillators labeled "Voltage Controlled Oscillator No. 1" (VCO-1) 
and "Voltage Controlled Oscillator No. 2" (VCO-2). Focus your atten- 
tion on VCO-1 (second panel from the left, Figure 2.2.1). 


Experiment 1: Waveforms 

VCO-1 generates three kinds of waveforms: square, pulse, and saw- 
tooth. The selector switch below the blue VCO-1 attenuator on the 
Audio Mixer section of the control panel chooses which VCO-1 wave- 
form will be heard. This switch should be on the sawtooth wave, as 
per the general patch. Gradually open this attenuator until you reach 
the top, and then close it in the same manner. Next, move the switch 
down to the square-wave setting and raise and lower the attenuator 
as you did before. Can you hear the difference in the two waveforms? 
Raise the attenuator again, and we'll look at some other controls that 
influence the sound of VCO-1. 

VCO - 1 







Figure 2.2.2. Block diagram of Figure 2.2.1. 


* I *.. 

X |£S 1 1 r- 


x a £ 



O cr 


z#— Isms? 

1 1 

1 1 


ft£X3 : 3 




4— > 

153 i 

* 5 1 r 


m * 


uj a o 



.. csi 
o n 

58 | s 


Experiment 2: Tuning Controls (VCO-1 ) 

The tuning controls (blue) for VCO-1 are found in the second panel 
from the left under the work "Frequency." Study Figure 2.2.3. 
Now, experiment with the coarse-tuning slider. What happens when 
you raise and lower the slider? Listen carefully while you move the 
slider through its entire range. Turn your attention to the fine-tuning 
slider and go through the same steps. Do the sliders have the same 
basic function? If so, which of the sliders is more sensitive? The 
answer to these questions is found in the following rule: As you 
raise the tuning sliders, the pitch (frequency) goes up; lowering the 
sliders brings the pitch (frequency) down. And "coarse" tune is the 
most sensitive control. 





How to tune VCO-1. Set the "Fine" control at center. Tune to approximately the 
desired pitch with the "Coarse" control. Then tune exactly with the "Fine" control. 

-I — r's 

*5£> 1- 


° z s 
> ol 


n t ,r 

— .; ; ; ; » — 1 1 |lsl 

E 1 C 

f < 

Si i 1 


8 6 


SB » 
/ > 

s * 

I s 


=*1 smi^ 




Experiment 3: LF Range 

Move the Audio switch shown in Figure 2.2.5 to the down position 
(LF KYBD OFF). Move the blue tuning slider all the way up - you'll 
hear the clicks become faster. The reason we hear separate clicks/events 
in the low-frequency range is due, of course, to the limited range of 
frequencies humans perceive as having pitch. We only perceive physical 
vibrations as pitch between 20 Hz and 20 KHz. Later on, you will 
learn how to employ the LF range of VCO-1 as a control device. 
For now, return the selector switch to the Audio Keyboard position 


Low Frequency 



Figure 2.2.4. Block diagram for Figure 2.2.5. 


t t 

O z o 

=»1 5 CON 

II I L„Ol 

1 1 

* in 








Figure 2.2.6. Block diagram of Figure 2.2.7. 

Experiment 4: Transpose Switch and Keyboard 

Set the controls of the Odyssey to conform to the general patch 
in Figure 2.1.1 with one exception — raise the blue VCO-1 attenuator 
in the Audio Mixer. Also, locate the Transpose switch found in the 
first panel on the left. Make sure the switch is in the middle position 
as shown in Figure 2.2.7. Now we'll be using the keyboard of the 
Odyssey for the first time. This keyboard looks like a section of the 
keyboard of a piano. A more technical explanation of how the key- 
board controls the various functions of the Odyssey will be presented 
later in this section. 



j w u < 

I Ifcri 




> o < 





j C j 

> o 



1 I I I 


O tr 


If mis 



— ^pdli 

j *? — I — £ 4 

j £ si — LJ« £ 




' It 


f j 




» d 




I CO/ 



Hold down "C and try all three positions of the Transpose switch. 

Figure 2.2.8. Transpose switch positions. 

Press and hold down one key on the keyboard, as shown in Figure 
2.2.8. While holding the key down,move the Transpose switch to the 
"down" position. What happened to the pitch? Do the same thing, 
only move the Transpose switch to the "up" position. Did you again 
perceive a change in pitch? Figure 2.2.9 uses musical notation to show 
what happened. As you can see, the Transpose switch shifts the entire 
keyboard control up or down by two octaves. In this way the keyboard 
range is extended to a full seven octaves. 








Figure 2.2.9. Transpose switch musical effect. 

Next, press individual keys on the keyboard. Listen to what 
happens as you play keys going to the right, and then as you play 
keys going to the left. To the left, the pitch goes down. To the right, 
the pitch goes up. 

The keyboard is actually an extremely precise voltage controller 
which determines the pitch of the oscillators by providing a control 
signal which is fed to the oscillators. The keyboard control voltage 
will be discussed in greater detail in this section under "Controllers." 


Lesson 3: Pulse Width 
What Pulse Width Is 

A pulse wave is a waveform that has only rectangular corners. A 

pulse wave can look like this (a) H 11 f l like this (b) JT Tl Fl I 

or like this (c) J~~L PL. T"LT . A pulse wave is described by the 
relative widths of the high and low portions of the waveform. If, 
as in the first example, the pulse wave is in the "high" part of its 
cycle for only a very short time, it is said to be a "narrow pulse 
wave" and will have a very bright, nasal sound. A wider pulse wave, 
(b) above, has a fuller sound. Example (c) is a special case of the 
pulse wave — the square wave. It is called a square wave because the 
"high" part of the wave is exactly the same length as the "low" part 
of the wave. The square wave has its own distinct clarinetlike sound. 
In fact every different width of pulse wave has it's own unique timbre. 

One interesting fact about pulse waves is that two pulse waves that 
are upside-down copies of one another will sound exactly the same. 
For instance, this pulse wave (d) f l fl fl I " will sound exactly the 
same as this pulse wave (e) y [j JJ |_ • Consequently, on a syn- 
thesizer it is only necessary to be able to create half of all the possible 
pulse widths, since the other half (where the high part is longer than 
the low part) sounds the same. 

Both of the VCOs on the Odyssey are equipped with controls labeled 
"Width." When a "Width" control is in the "down" position, the 
pulse width is exactly 50%, making it a square wave. As the "Width" 
slider is raised, the pulse gets narrower and narrower until it looks 
like waveform (a). 





j r 

Figure 2.3.1. The pulse "width" control varies the shape of the pulse waveform. 

Experiment 1: Changing Pulse Width 

Set the controls as pictured in Figure 2.3.2; then raise and lower 
the blue Pulse-Width slider through its entire range. As the slider 
moves, listen carefully to the changes in timbre/tone color. Here is 
what happens when the Pulse-Width slider is raised from its lowest 



* < . 



Figure 2.3.3. Diagrams for pulse waveforms at 50%, 25%, and 2%. 

When the waveform 
is even like this, it's 
called a square wave. 

A 50% Pulse Wave 

|4- 100% width ->! 

' 50% fc L 50%J 
~ up "*^down^ 

About halfway up, 
the waveform looks 
like this. 

A 25% Pulse Wave 

J4- 100% width 
i i i 

25% 75% 
up down 

When the slider is 
all the way up, the 
waveform looks 
like this. 

A 2% Pulse Wave 

4100% width 

jL ^ 

2% 98% 
up down 

Return the slider to the 50% position before continuing. 


Experiment 2: Pulse-Width Modulation 

Pulse-width modulation simply means changing the pulse width, 
either manually as you did in Experiment 1, or by using a control 
voltage, as we will now do. Locate the pink Pulse-Width Modulation 
slider shown in Figure 2.3.4, and set the controls accordingly. 
With the pink Pulse-Width Modulation slider raised, the rate of change 
will now be determined by the pink LFO Freq slide control in Panel 
4 (Figure 2.3.4). Set the LFO to a very low frequency and you'll hear 
the tone getting very nasal, then very hollow. Listen to various LFO 
settings. This, of course, is an example of voltage control; the LFO 
is controlling the timbre of the output of VCO-1 automatically. You 
can do the same thing by raising and lowering the Pulse-Width slider 
manually. Try it. 

Experiment 3: Pulse-Width Modulation (VCO-2) 

Set the controls under the Audio Mixer as shown on Figure 2.3.6 
and perform the entire preceding experiment using VCO-2 instead of 
VCO-1. When you've completed the experiment, return the Pulse- 
Width controls to the closed position. 

VCO - 1 
Cont. PWM 







Figure 2.3.5. Block diagram of Figure 2.3.4. 





e 5 y 


mod _n_n_ jut. 

Figure 2.3.6. 
Pulse- width modulation 
experiment (VCO-2). 




• HI mil 

?8 t 

4=*] 5 mi/ 








Lesson 4: Frequency Modulation 

Remember that systematically changing the frequency of a wave- 
form by employing the pattern of a second waveform is called 
frequency modulation. Refresh your memory on the concept of 
frequency modulation by turning back to Part I of the text. Set the 
Odyssey controls as shown in Figure 2.4.1 on the next page and 
you're ready to try some examples. 







Figure 2.4.2. Block diagram of Figure 2.4.1. 

Figure 2.4.3. LFO switch (VCO-1). 


I B 



Experiment 1: Vibrato and Trill 

Locate the pink FM (frequency modulation) attenuator (second 
panel from the left). For now, just use the first (left) slider, as shown 
in Figure 2.4.3, and experiment with it in several positions. You should 
hear a vibrato. Directly under the slider, you'll find a two-position 
LFO switch; move it down and you'll hear a trill. As you raise the pink 
slider, the two pitches of the trill become farther apart; lowering the 
slider brings the pitches closer together. Experiment with the blue 
Pulse-Width slider, the pink LFO Freq slider, and the VCO-1 selector in 
Panel 5, to create additional effects. 

r ~..w..J.>,..f::; 



uj u c 

o J a 


la j 

2 o Si 
j !j 
9 z c 

t— i r- 

• !la=EDs||s| 

O <r 


ir in 

£ o 

I J 


. | ^ — , 


3 3 £ 


! * azi/ 


I c 

1 1 s *• 

1 I <; ; # 


gg H — 

uj "J O 



li s i i i W 





















2 (J- 

o 1 
< s 

ft UJ ► i 

w a < >— 




z* ["gj p=gj 

L * j 


UJ "j 

J !_ 

S r- 

O cr 
Q w 
3 £ 


:*H« 5 en? 

— ||a=ni||s| 

• -li mi! 


—<::::» 1 


_ j — i — i — i — jm* 1 


1 1 


lu j O 

O J 5 

2 2 ^ . 

j EE d 

o x o 

S$P3 a 




a lam 

;rjr * 


















Experiment 2: Voltage Control 

The second frequency-modulation attenuator in Panel 2 (yellow) 
permits you to voltage-control the pitch (frequency) of VCO-1 and 
VCO-2 with either the Sample and Hold function (Panel 4) or the 
ADSR envelope generator (Panel 6). Be sure that the controls of the 
Odyssey are set as shown in Figure 2.4.4. If they have been set cor- 
rectly, you should hear the sound of the VCO-1 sawtooth wave. Now 
raise the second slider (the white one, second from the left) on the 
S/H Mixer panel (Panel 4). After doing this, raise the frequency- 
modulation slider (yellow, on VCO-1) located above the S/H- ADSR 
selector switch. The result should be a series of rapidly changing 
pitches, the speed of which can be controlled by the LFO Freq slider. 
Vary the speed (frequency) of the LFO. Then perform the same exper- 
iment using VCO-2, setting the controls under the audio mixer as 
shown on Figure 2.4.5. 



I I I 
> 1> ♦ 

* # -u- 



I H B 


Figure 2.4.5. Audio Mixer controls 

for using VCO-2. Figure 2.4.7. Same as Figure 2.4.5. 

Experiment 3: 

After experimenting with the Sample and Hold function, return 
to the settings shown in Figure 2.4.4 and push the two-position selec- 
tor switch in Panel 2 down from S/H to ADSR (Figure 2.4.6). The 
frequency (pitch) of VCO-1 may now be controlled by the ADSR 
envelope generator. Raise the yellow slider on VCO-1 again and vary 


S/H . 

B 1 


Figure 2.4.6. 
selector switch (VCO-1). 





B H I 

MOD _n_n_ J-LTL 


the attack, decay, sustain, and release settings to find out how the 
pitch of the oscillator is affected. Playing any note on the keyboard 
will trigger the envelope generator. 

Experiment 4: Voltage Control (VCO-2) 

Change the controls under Audio Mixer so that you can listen to 
the output of VCO-2 (Figure 2.4.7). Repeat Experiment 3, now using 
the controls on VCO-2. 

Each of the above procedures provides a method of voltage-control- 
ling the sound sources (oscillators). The precise means by which each 
function — the Sample and Hold and the ADSR — exerts its control 
will be discussed later under "Controllers." 

Lesson 5: Synchronization 

Synchronization of the two oscillators (VCO-1 and VCO-2) is en- 
abled by the selector switch shown in Figure 2.5.1. When this switch 
is "on" (down position), the audio signal from VCO-2 is forced to 
conform to the frequency of VCO-1. For a demonstration of synchron- 
ization, set the controls to match Figure 2.5.2. 



Figure 2.5.1. Synchronization switch. 


'II r 



a H oc 


a oc 

5 x 
3 3 


J 1 I L_ 


ii ii< 

I • -i 




S * S! 

*1 sm|< 


a ui 




o z o 

=#1 scn|^ 



3» I A 


With the Sync switch on, slowly move the green coarse-tuning slider 
of VCO-2 through its entire range. Notice the change in timbre. Move 
the pitch of VCO-1. Notice that the pitch of VCO-2 changes with the 
pitch of VCO-1. At certain settings of VCO-2 pitch, the pitch of 
VCO-1 disappears, and you only hear a high-pitched harmonic of 
VCO-1. The disappearance of the pitch of VCO-1 results because 
you tuned VCO-2 to an exact harmonic of VCO-1. Incidentally, if 
the frequency setting of VCO-2 is below that of VCO-1, the volume 
becomes softer and the timbre does not change. As the green VCO-2 
slider is raised higher than the blue VCO-1, the timbre of VCO-2 
does change. Experiment again, changing the frequency of VCO-1 
while leaving VCO-2 in the middle of its range. The basis for some 
exciting sounds will be found in this operation. 

Experiment 1: How Synchronization Works 

Let's look at some diagrams which will show graphically the syn- 
chronization of VCO-2 to VCO-1. Start by listening to the sawtooth 
outputs of both audio oscillators with the Sync switch in the "off" 
position (Figure 2.5.3). Set VCO-2 (green) so that it has a slightly 
higher pitch than VCO-1. Figure 2.5.4 shows the difference in wave- 
forms due to the higher pitch (frequency) of VCO-2. Now turn the 
Sync on. What happens to the VCO-2 waveform is shown in Figure 

When VCO-2 sounds with the Sync switch on, VCO-1 interrupts 
or "resets" the signal from VCO-2 when VCO-1 begins each new 
cycle. Figure 2.5.5 may help to explain. Again, move the coarse-tuning 

VCO - 1, LOWER PITCH (Frequency) 

VCO - 2, HIGHER PITCH (Frequency) 

Figure 2.5.4. Sawtooth waveforms out of synchronization. Sync switch, "off." 


VCO - 1 

VCO - 2 I I 


Figure 2.5.5. Comparison of sawtooth waveforms out of and in synchronization. 

slider on VCO-2 through its entire range with the Sync switch on, 
going from low to high. Listen carefully as the fundamental of VCO-1 
gets weaker. Raising the slider will enable you to center in on the 
harmonics of VCO-1 with the fundamental eventually fading com- 
pletely, then returning as the VCO-2 slider is moved higher. Don't 
hesitate to use the frequency-modulation and pulse-width controls 
to see the various effects which you can create, always being sure to 
picture mentally what the waveform changes would look like on paper. 

Lesson 6: Tuning the Oscillators 

Open the blue VCO-1 and green VCO-2 attenuators into the Audio 
Mixer as shown in Figure 2.6.1. 

Experiment 1: Unison 

Position the blue coarse-frequency control slider of VCO-1 to a 
medium pitch. Then manipulate the coarse-tuning slider of VCO-2 
until it sounds close to the pitch of VCO-1. You'll perceive a contin- 
uous series of rapid beats or waverings as you near the pitch of VCO-1. 
Use the fine-tuning control, continuing to make adjustments until 


a < 

5 i 9 


Ki ■:: *» S 



1 1 1 r- 


< o t 


| 8 js 

x a a; 

5 g £ 


o cc 

3 2 



■ I 


1 I 1 I 


u ^ 


t ♦ 

a d s 


=*1 scq^ 





vco - 





Figure 2.6.2. Block diagram of Figure 2.6.1. 

the beats slow down and finally stop. When the beats terminate, the 
oscillators are tuned to exactly the same pitch — they are now "in 
unison." Following the same procedure, tune the oscillators to many 
different pitches, both low and high. 

Experiment 2: Interval Tuning 

If you wish to try a more challenging task, tune the oscillators to 
various intervals such as octaves, fifths, and thirds. (You'll find this 
may take a bit more practice.) 

Section 3: Signal Modifiers 

Now that you have examined various sources of raw-signal produc- 
tion, it's time to explore those components of the Odyssey which 
fashion the raw material/signals into an audibly useful form. These 
components, called signal modifiers, are: (1) the voltage-controlled 
filter (VCF), (2) high-pass filter (HPF), (3) voltage-controlled amplifier 
(VCA), and (4) the ring modulator — all on Panel 5. One of the four 
signal modifiers — the VCF — has the dual capacity of both a signal 
source and a signal modifier. This dual function will be discussed in 
Lesson 3. Let's begin by learning about the first and most important 
of the Odyssey's signal-modifying devices. 


Lesson 1: Voltage-Controlled Filter (VCF) 

Find the VCF components shown in Figure 3.1.1 — three attenua- 
tors (black, yellow, and red) located below the words "Voltage Con- 
trolled Filter" and two sliders (black) above and to the left labeled 
VCF Freq and VCF Resonance. The Voltage-Controlled Filter (VCF) 
is a low-pass band-pass filter which can be controlled in two ways: 
either by a voltage from another circuit or manually, using the "VCF 
Freq" and "Resonance" controls. Keep in mind that filter names such 
as "low-pass" are descriptive in nature. 








0 B B 


MIXER AA/ J^~\. 

Figure 3.1.1. Location of Voltage Controlled Filter controls. 


Raise the white noise slider shown in Figure 3.1.2 (Audio Mixer) 
and experiment with different positions of the black VCF frequency 
slider. Next, move this slider to the top of its range. Then gradually 
lower it all the way through the entire VCF Freq range. You'll hear 
the high frequencies of the noise signal weaken and disappear, then 
the middle frequencies, and finally the low frequencies will also vanish. 






Figure 3.1.3. Block diagram of Figure 3.1.2. 

Experiment 2: VCF Resonance Slider 

The VCF Resonance slider is located directly to the right of the 
VCF Freq (Figure 3.1.4). As the Resonance slider is advanced, the VCF 
changes from a low-pass filter into a band-pass filter. As the Resonance 
slider is raised, you'll hear the low frequencies fade as the band of 
frequencies which is allowed to pass through the filter becomes 
narrower. Use noise as the signal source, and experiment with various 
positions of the Resonance slider. 

Experiment 3: 

Repeat Experiment 1, but with the Resonance slider halfway up 
(Figure 3.1.5). 





Figure 3.1.4. VCF Resonance control. Figure 3.1.5. VCF Resonance control. 





Experiment 4: VCF Resonance Control of VCO-1 and VCO-2 

Perform the preceding experiment using the sawtooth wave VCO-1 
as the signal source. Listen as the pass-band — those frequencies allow- 
ed to pass — sweeps through the harmonics of VCO-1 as you raise 
and lower the VCF Freq control. Experiment with different settings 
of the Resonance control. 

Experiment 5: ADSR Control of VCF 

First locate the ADSR Envelope Generator controls in the lower half 
of the last panel on the right (Figure 3.1.6). Carefully set the controls 
as pictured, and you'll be ready for an example of how the VCF may 
be voltage-controlled by the ADSR Envelope Generator. 

Begin by moving the red ADSR attack slider to various positions. 
The change in sound is due to the "automatic" voltage control of the 
VCF by the ADSR Envelope Generator. Return the attack slider 
control to its lowest position and experiment with the decay, sustain, 
and release sliders. The speed of repetition of the sound is controlled 
by the pink LFO Freq slider in Panel 4. 







Figure 3.1.7. Block diagram of Figure 3.1.6. 




2 i 






*BeS 1 1 r- 

I < " 

I o. c 

o I 
a £ 

i 1 r- 

• -lien* 


I I 3 

« •-: 

>• M — : — i 5 



j i i i 


O j H 

o £ 5 

8 © 

*1 icr^i 







Try different settings of the red attenuator which controls the 
amount of ADSR signal to pass into the filter. This attenuator is 
located just under the box labeled "High Pass Filter" and just over 
the ADSR/AR switch. Notice that this attenuator determines the 
"brightness" of the sound. The reason it does this is because the 
voltage which is produced by the ADSR Envelope Generator is being 
used to "open" the VCF, which in turn lets the high-frequency (bright) 
sounds pass through. The attenuator will reduce the amount of ADSR 
signal which passes to the filter's control input and will determine 
how far the filter can "open." The lower the setting of this attenuator, 
the less ADSR signal is allowed to pass; consequently, the less the 
VCF will respond to the output of the ADSR. 

Before going on to Experiment 6, be sure to review the information 
presented in Part I on the function of the envelope generator. We 
will discuss both the AR and ADSR generators again under "Con- 
trollers"; however, it is important to understand here that by manip- 
ulating the controls of the ADSR you are shaping an aperiodic 
waveform that is opening and closing the VCF in accordance with the 
settings of the ADSR controls. By voltage-controlling the filter in 
this manner, you have a precise and extremely sensitive means by 
which you can control the timbre of any sound. 

Experiment 6: Filter Tremolo 

Tremolo is a form of timbre modulation. Set the controls to the 
positions illustrated in Figure 3.1.8 and you'll hear an example of 
tremolo. The sine wave output of the LFO is being used to open 
and close the VCF automatically, as if you were moving the pink 
VCF Freq control back and forth by hand. Try changing the setting 
of the LFO Freq control and notice how the speed of the tremolo 
changes. Vary the position of the yellow slider over the S/H-LFO 
switch and notice how this attenuator varies the depth of the tremolo, 
in much the same way the red slider next to it controlled the amount 
of signal from the ADSR that reached the filter in the previous experi- 
ment. Listen to each of the three sound sources one at a time by 
raising the white, blue, and green sliders under the Audio Mixer box. 
Change the setting of the VCF Freq control and notice how this 
control still determines the overall brightness of the sound. 

II gV- 
$11 \y> 




> o < 

1 * s 


j 5 < 


u. O 






O z u 

1 \ mi 

i? 1183311 

• II mil 



VCO- 1 



Figure 3.1.9. Block digram of Figure 3.1.8. 


Experiment 7: Voltage-Controlled Filtering of the Noise- Generator 

In the beginning of this section you created the sound of surf by 
manually controlling the output of the noise generator. Now you 
can use a voltage control to produce the same sound. It will be sur- 
prisingly more realistic this time. For a demonstration, set the controls 
according to Figure 3.1.10. Note the position of the LFO Freq slider. 
After listening to these sounds, do you agree that voltage control 
has many advantages over manual control? 

Lesson 2: High-Pass Filter 

Another component which functions as a signal modifier is the 
High-Pass Filter (HPF). Figure 3.2.1 shows the location to the left 
of the VCA Gain slider in Panel 5, labeled "HPF Cutoff Freq." 






Figure 3.2.1. Location of High-Pass Filter. 


i ifpdp 



5 SjS 
5~~ * | ^ 

x a. £ 



9 £ 




=b— J! Kill 



I S 


£ '■> \ - 

I . S I" 

* K : <•> 


3P I 









The high-pass filter does exactly what its name implies — it lets the 
high frequencies pass through while filtering out the low frequencies. 
This particular filter is not voltage-controlled, and can only be adjusted 
manually with the black slider. The HPF is useful in removing the 
"boomy" quality from lower pitches, and also aids in the texturing 
of certain instrumental sounds. Set the controls as illustrated in 
Figure 3.2.2. 

Experiment 1: Filtering VCO-1 with HPF 

Move the HPF Cutoff Freq slider through its entire range. Notice 
how the low frequencies weaken and finally disappear as the slider 
reaches the top. 

Experiment 2: Filtering White and Pink Noise 

Use the same patch shown in Figure 3.2.2; this time, however, 
lower the blue slider over the VCO-1 selector switch (under the Audio 
Mixer box) and raise the white slider to hear the noise generator. 
Again raise and lower the HPF Cutoff Freq slider to observe the effect 
of the HPF on noise. Filter both white and pink noise. Remember 
that pink noise is produced by moving the selector switch of the noise 
generator to the "down" position. Of course, the HPF has the same 
effect on both periodic and aperiodic waveforms: it can remove low 



Figure 3.2.3. Block diagram of Figure 3.2.2. 


Jt% It: 

SJ « 

■ t i 


_ 111 




* o 




• II mil 

# " — 


gg r 1 1 1 

9 ' S— 1 P- 

^ ! 


2 5 [f^ 
1 . 1 , 


J J J 

"i 1 r- 











1! U 

1 1 


Lesson 3: VCF as a Tone Generator 

At the start of this section on signal modifiers you learned that 
one component, the VCF, had the dual capability of both signal modi- 
fier and signal source. To turn the filter into a sound source, raise 
the Resonance slider all the way up (Figure 3.3.1). Focus your atten- 
tion on the words Self Osc located to the right of the Resonance 


Figure 3.3.1. Using the VCF as a VCO. 

Experiment 1: VCF Resonance in Self-Oscillation 

Match the controls of the Odyssey to those of Figure 3.3.2. 
In this position, the VCF is no longer a signal modifier; it's acting as 
an oscillator producing a raw signal. As you can see, all the attenua- 
tors in the Audio Mixer are down, yet a sound is being produced. 
The waveform you are hearing is a pure sine wave. Raising the black 
attenuator over KYBD CV switch allows the keyboard control volt- 
age (KYBD CV) to alter the frequency of this output. Note that the 
pitch of the tone produced by the self-oscillating filter can be changed 
both by the keyboard and by the VCF Freq control. There is really 
little difference between the operation of the filter when used as an 
oscillator (when the Resonance control is all the way up to Self- 
Oscillate) and either of the VCOs. You can even add vibrato to the 
pitch from the filter by raising the yellow slider which feeds the LFO 
/"\./ = X^ / into the filter. Be sure that the S/H-LFO switch is set to 
the LFO position if you want to try this 










! II 







Figure 3.3.3. Block diagram of Figure 3.3.2. 

Experiment 2: Playing Microtonal Scales on the VCF 

The black slider under the VCF box allows the keyboard control 
voltage to control the filter, as we demonstrated in the previous ex- 
periment. Lowering this slider allows you to attenuate the keyboard 
control voltage before it reaches the filter and thereby permits you 
to reduce the pitch change created by the keyboard. Set this slider 
approximately halfway up, as shown in Figure 3.3.4. Note that the 


H H 

■ ■ >i J S i . 









Figure 3.3.4. Creating microtonal scales. 


range of the keyboard has been cut in half and playing an octave on 
the keyboard results in a pitch change of only half an octave. Similarly, 
every semitone (half tone) on the keyboard will now equal a quarter 
tone. Such a scale tuning is called a "quarter-tone scale." As you 
bring this attenuator down further, the pitches between the keys will 
become closer. Try setting the slider so that 3 octaves on the keyboard 
produces a pitch change of one octave. 


















Keyboard has no effect 
on pitch if slider is 
at minimum. 

One octave on keyboard 
= l A octave musically 

One octave on keyboard 
= one octave musically 

Figure 3.3.5. Different microtonal scales. 


Lesson 4: The Ring Modulator 

Another signal modifier which you'll learn to use is the Ring Mod- 
ulator. The following diagram (Figure 3.4.1) will help you to under- 
stand this important component better. Study this "assembly line" 
illustration for a moment. First of all, the Ring Modulator has no 
controls of its own, and, as you can see from Figure 3.4.1, utilizes 
the outputs of VCO-1 and VCO-2. From the outputs of these two 
signal sources, the Ring Modulator produces a single complex output 
signal which contains all the sums and differences of the two oscil- 
lator frequencies. The output (raw sound) created by the Ring Mod- 
ulator is entirely dependent upon the tuning of VCO-1 and VCO-2, 
and to a lesser degree upon the pulse-width settings of each signal 
source. If the two inputs are at two different frequencies not related by 
simple harmonic ratios, the Ring Modulator output will have a very 
complex characteristic since it contains high-frequency components 
which are not harmonics of either fundamental frequency. Locate the 
Ring Modulator switch as shown in Figure 3.4.2. Be sure the switch is 
in the "down" position. Set the other controls as shown. 











Figure 3.4.1. Block diagram of Ring Modulator function. 

Experiment 1: Tuning for Effect 

Raise the white Ring Mod attenuator located in the Audio Mixer 
section. With the Ring Mod attenuator in its highest position, begin 
moving the tuning sliders of the VCOs to different positions. As the 
tuning controls are changed, you'll hear the sound change. Change 
the position of the Pulse-Width sliders on VCO-1 and VCO-2. Ob- 
serve the effect. 


?M It 

f c, 

2 < oi 

1 C 


Ui uj 
© T E 


aid— !- 

2 a 
— l- 


C cr 





J L 

ui a o 

:#] §£33||sf 


( i i i 

I 3 <£z 

I 5 




5| d ' 

=#1 smf^ 


if 11 [ If 











Experiment 2: Other Controls that Affect the Ring Modulator 

Let's attempt to create a gonglike sound by employing the Ring 
Modulator and ADSR Envelope Generator. Again find the ADSR 
controls in the bottom half of Panel 6. Figure 3.4.3 illustrates. Set 
all the controls as shown. Try the following and observe the effect: 

1. Change the positions of the pink and yellow VCO-1 and VCO-2 
tuning controls and Pulse Width controls. 

2. Change the position of the red attenuator under the High Pass 
Filter box. Note that this control affects the brightness of the sound. 
Recall that we experimented with this control in our discussion of 
the VCF. 

3. Change the position of the VCF Freq control and note that it 
too influences the brightness of the sound. 

4. Change the setting of the VCF Resonance control and listen to 
the effect that this control has on the sound. Remember what effect 
this control had on sounds in previous experiments. 

5. Experiment with different settings of the four red ADSR En- 
velope Generator controls. 

6. Try changing the positions of the pulse-width and frequency- 
modulation sliders on the VCOs to create additional sounds. 

7. When the sync switch is in the "on" position, VCO-2 will be 
"locked" to a harmonic of VCO-1 even if VCO-2 is deliberately detuned 
when the sync switch is in the "sync off" position. Try using the sync 
switch to change rapidly from a gong tone to a bright "harmonic" tone. 



VCO - 2 






Figure 3.4.4. Block diagram of Figure 3.4.3. 




: }f r 

$11 If* 




33 §<r 

I 1 !* 

r S 5 

i ft* 


T 1 1— 

KM 4 



# — llimig 


S s -i 
e 5 <= 







• seen 



Lesson 5 : The Voltage-Controlled Amplifier 

The last signal modifier to be discussed is the Voltage-Controlled 
Amplifier (VCA). This component determines the final volume or 
amplitude of the output before it leaves the Odyssey. Find the VCA 
Gain slider and its companion voltage-control attenuator (red) located 
directly below it in Panel 5 (Figure 3.5.1). 

The function of the VCA is to control the amount of signal passing 
from the other signal modifiers to the output. It is essentially a volume 
control that can be operated either manually (using the VCA Gain 
control) or automatically by signals from the AR Envelope Generator 
or the ADSR Envelope Generator fed in through the red attenuator 
located under the Voltage Controlled Amplifier box. 

Raising the VCA Gain control will have the effect of permitting 
any signal passing through the other signal modifiers to reach the 
output of the synthesizer. The higher up the VCA Gain control is set, 
the greater the amplitude this signal will have at the output of the 
synthesizer, and, consequently, the louder the sound that will come 
out of the speaker you are using. 

If a continuous signal exists at the input of the VCA which you 
wish to turn on and off when you are playing on the keyboard, for 
instance, you would normally leave the VCA Gain control all the way 
down and use the signal from the Envelope Generators to open and 
close the VCA. In this case, the red attenuator under the VCA box 
will control the volume of the over-all output of the Odyssey. The 
switch below the red slider will select control of the VCA by either 
the AR or ADSR Envelope Generators. Position the controls as pic- 
tured in Figure 3.5.2 and you'll be ready to demonstrate these 



• ! 






I r~ 

& £• 

■V. v 

if 3 
$ ~ 2 ■ 

; A' * 


I 1 , jL 

I ■ 

I 5 

j i 


if mi 

yr * i 

i or 


• 4 


• -II mis 

* I 








Experiment 1: VCA Gain 

With the controls set as shown in Figure 3.5.2, you should not be 
hearing any sound. Gradually raise the VCA Gain attenuator to its 
highest position and the volume will increase accordingly. Lower 
the slider again. Note that you are only changing the amplitude 
(volume, in this instance) of the audio waveform you're hearing. You 
are not altering the harmonic content of the sound, as you would if 
you closed the VCF Freq slider. Prove this to yourself by doing just 
that. Raise the VCA Gain slider. Now, as you raise the VCF Freq 
slider, you'll hear the complexity of the audio signal increase. By the 
time you reach the top of the slider's range, you will hear the full 
harmonic content of the waveform. Closing it again will have the 
reverse effect. 

The point here is that the use of the VCA, its functions and capa- 
bilities, cannot be freely interchanged for the use of the VCF. Each 
has its own particular functions, which when used together will permit 
you to create effects that neither could produce alone. The two fol- 
lowing experiments demonstrate this. 

Experiment 2: ADSR-Controlling the VCA 

Set the controls of the Odyssey as shown in Figure 3.5.3. By open- 
ing the VCA Gain slider and closing the ADSR control attenuator into 
the VCA, you can prove that with the filter in oscillation you would 
have no control over the attack and/or duration of the audio signal 
produced. Then return to the position of Figure 3.5.3. Using the Key- 
board Control Voltage to determine the pitch and the ADSR to 
control the VCA (providing an envelope to control the amplitude of 
the waveform), you can now "play" the filter with complete control. 



I o 11 

r 5 * t — 

< 1 9 — 

32 gS= 

>3 ( r = r : 




• mig 

• || mil 



i aim 

1 1 1! 

« 1 f 

II I* 

p x. 5 

1 CIj|< 

Iff ESQ 








Figure 3.5.4. Block diagram of Figure 3.5.3. 

Experiment 3: ADSR-Controlling the VCA and the VCF 

The ADSR control of the VCA will be equally useful even when 
the filter is not in oscillation. There are times when you will want to 
leave the filter partially open, without hearing the continuous "bleed" 
of the frequencies not being filtered (Figure 3.5.5, next page). 

VCO - 1 







Figure 3.5.6. Block diagram of Figure 3.5.5. 


Play a series of notes on the Keyboard. You will find that even 
with the long release time on the ADSR, the sound will die away 
completely. By opening the VCA Gain slider again, however, you will 
find that the filter is really partially open — without the VCA, such a 
filter setting would not produce an effect that would be nearly as 
pleasing, or as useful musically. 

This partial block diagram (Figure 3.5.7) of the patch shown in 
Figure 3.5.5 shows the audio signal from VCO-1 passing through the 
VCF and VCA (the signal also passes through the HPF but is not 
affected because the HPF control is all the way down) and the control 
signals for the VCF and VCA both coming from the ADSR Envelope 






Figure 3.5.7. Partial block diagram of Figure 3.5.5. 

Experiment 4: AR Control of VCA While ADSR-Controlling VCF 

At times, you will want to have the option of controlling the VCF 
and the VCA with differently shaped envelopes. You'll note that both 
the VCF and VCA have attenuators and two-position switches (the 
one farthest right on Panel 5) permitting you to control either, or 
both, the VCA and VCF with the AR or ADSR generators. 


sis a j 

J2 I 



[ ■ 

rTn-/ , 

Ui w 

o d « 
2 o £ 




z#Hl mis 


s I CZ 

1 1 ?l 



i 1 1 

I sis 


ES t 

2 § li 



=#1 iSP3|if 










Si S CO I 





Note in Figure 3.5.8 that the VCF is being controlled by the ADSR 
and the VCA by the AR. After playing a few notes using the keyboard, 
close the red AR Release slider all the way. While the ADSR envelope 
controlling the filter is still programmed for a long release, the AR 
controlling the VCA is overriding the release time by simply closing 
the VCA — just as you would turn off the volume control of a stereo. 
Though a record album might still by playing, you would no longer 
hear its sound coming from your loudspeakers. 

Similarly, by slowing the attack of the AR, you can slow the attack 
of the sound, even though the filter is programmed by the ADSR for 
fast attack. Prove it to yourself by raising the AR attack slider. 

We can represent the signal flow in this fairly complex patch by 
the following block diagram. Study this diagram and see how every 
interconnection shown on the block diagram is also shown on the 
Odyssey front panel (Figure 3.5.9). 

You may have deduced by now that the VCA is working like a 
volume control that, when open, lets you hear what is coming out 
of the filter (VCF). Consequently, if you wanted to hear everything 
that was going on in the filter section (for instance, how the ADSR is 
opening and closing the filter), you must set the controls of the AR 
so that the VCA is "open" during the complete time of interest. In 
other words, if you had a very short attack time programmed on the 
ADSR, but a long attack time programmed on the AR, the ADSR 
would be finished with its attack (remember, the ADSR is opening 
the VCF) before the AR opens the VCA enough for you to hear 
what's going on in the filter. So if you want to hear that fast attack 
from the ADSR and VCF, you will have to open up the VCA quick- 
ly enough — and that means setting the attack time of the AR to be 
very fast also. 

Similarly, if you have programmed the ADSR for a long release 
time so that the VCF will close slowly, you must also set the AR for a 
long release time so that the VCA will stay open long enough for 
you to hear the filter's long release. 

Experiment on your own to establish in your mind the flexibility 
that this relationship between the VCF controlled by the ADSR and 
the VCA controlled by the AR can permit. You should also try revers- 
ing the position of the patch switches — control the VCF with the AR 
and the VCA with the ADSR. Note again that in either case you may 
also open the VCF Freq to any degree you desire, without continuous 
"bleed" of the unfiltered frequencies. Properly employed, based upon 
the understanding of the experiments above, the VCA becomes an 
extremely useful function in the electronic synthesis of numerous 
sounds and effects. 



lie • 



5 2 5 ll 




< o £ 


> o < 

> Q 

*Sp 1 

Ujg J I 1 L_ 

>£S 1 1 1- 



■ i 

=p — |la=i=E|si 

• ♦ : I : » Ht ■=> 

•HI mil 

j i 1 — u 





~l 1 — T 



•|1 C 





E 5 £ 



f 5 ^ 

— ■I ; ; ; # 1 

i—i 1 r 






% 1 

• 1 4 


Figure 3.5.9. Block diagram of Figure 3.5.8. 



Section 4: The Controllers 

Having now examined both the signal sources and the signal modi- 
fiers of the Odyssey, we come to the third general category of the 
synthesizer's functions: controllers. You will be working again with a 
number of voltage-controlled functions used earlier in Part II: the AR 
and ADSR generators, the LFO, the Sample and Hold, VCO-1 in its 
low-frequency range, and the keyboard. At this point, however, we 
shall explore more fully the precise means by which these controllers 
operate to produce some of the effects you've created. 

There are, of course, two methods of control: manual control and 
voltage control. When you use your hand to tune an oscillator, or to 
open the VCF or VCA, you are controlling these functions manually. 
The controller is your hand. However, each of these tasks can also be 
performed by a control voltage; hence, we have the voltage-controlled 
oscillators, filter, and amplifier. While your hand is a convenient con- 
trol device, if you think in terms of the great number of controlling 
functions that must be performed rapidly in order to create even a 
simple sound, you'll see that voltage control is essential. 

Consider, for example, what would be required to play a brief 
melody without the aid of voltage control. Certainly, it would be 
possible to play a relatively simple one by hand-manipulating the 
coarse-tuning slider of the oscillator being employed. Try it. You'll 
find, of course, that it requires some practice to "hit" the pitches 
(frequencies) required with any degree of accuracy. Moreover, you 
have no dynamic control (control over the loudness and softness), no 
expressive articulation of the individual notes (control of attack and 
duration), no control over the timbre (control of the harmonic con- 
tent or tone color), and no way to go from note to note without 
"sliding" across all frequencies between any two pitches. 

It becomes obvious, then, that we need the keyboard controller 
to provide accurate, instantaneous voltages for pitch determination; 
we require transient, or aperiodic, voltages that we can preshape and 
summon upon demand to control the amplitude, attack, and duration 
of various sounds; and finally, we often require voltages that extend 
our capacities even further by helping us control the controllers. 

The next series of experiments is designed to acquaint you with 
the controllers that you have at your command on the Odyssey. In 
each instance, you will be voltage-controlling at least one particular 
function, and in most cases, more than one. As you perform each 
experiment, try to picture mentally the effect the control voltage is 
having upon the function being controlled. Think of how you would 
have to manipulate the controls manually to create the same effect. 




When you can do this, you'll be well on your way to not only an 
understanding but a genuine appreciation of the virtually limitless 
potential for creative synthesis that voltage control can and does 

Lesson 1: The Envelope Generators 

The AR and ADSR generators, in the case of each controller we'll 
examine, produce control voltages which you can connect to other 
functions of the Odyssey by using the two-position patch switches 
and corresponding attenuators. In the case of the AR and/or ADSR 
generators, the connections shown in Figure 4.1.1 can be made. 

Note that there are eight possible connections — six for the ADSR 
and two for the AR. The ADSR can be used to control the frequency 
of both oscillators, the pulse width of both oscillators, the voltage- 
controlled filter, and the voltage-controlled amplifier. The AR can 
also be used to control both the VCF and the VCA. Furthermore, 
each envelope generator can control one or all of the functions to 
which it is connected simultaneously. You could conceivably use the 
ADSR to control frequency, pulse width, the VCF, and the VCA all 
at the same time. 

Let's begin by exploring again the most common use of the envelope 
generators — controlling the VCF and VCA. We will use many of 
the same patches used in our discussions of the VCF and VCA, but 
we'll look at them from the point of view of the Envelope Generator. 

Figure 4.1.2. AR envelope. 


Experiment 1: ADSR Control of the VCF and VCA 

Simply stated, the ADSR permits you to preshape an aperiodic 
waveform which will serve as a control voltage to open and close the 
VCF, VCA, or both. This permits you to control the way a sound 
begins, how long it lasts and how it sounds (the timbre) while it lasts, 
and finally, how long it takes to fade away. You shape this control 
voltage through the use of the four red control sliders in Panel 6. 

1. Attack How a sound begins. Most sounds have their own char- 
acteristic attack. A piano produces a sound having a relatively fast, 
immediate attack when a key is struck. A flute has a less sudden attack. 
A violin also has a gradual attack when a bow is drawn across the 

2. Decay. After the attack, a natural fading away of the sound 
occurs. A piano note begins to decay immediately; a guitar note decays 
more slowly at first. An organ note does not decay at all until the 
key being played is released. 

3. Sustain. The harmonic content of the audio wave varies from 
stage to stage, from attack to release. The sustain slider permits you 
to control: (1) whether the initial decay will be halted before the 
final release, and (2) if so, what the harmonic content of the waveform 
will be during this period. For instance, a trumpet sound normally 
has a slight "overshoot" or emphasized attack. Expressed another way, 
the sound begins with a burst which drops back to some other level 
for the duration of the note. The Sustain control will affect the level 
to which the sound drops back after the initial attack on the sound. 
If the Sustain control is set to minimum, the sound will die out com- 
pletely after the attack. If the Sustain control is set to maximum, 
there will be no "overshoot" and the sound will be at its loudest as 
long as you hold down a note. 

4. Release. The final fade-time of the sound is controlled by the 
release slider. As mentioned above, a guitar string would have a rather 
long release time since it requires some time to cease vibrating, unless, 
of course, the string is muffled by the player's hand. An organ tone, 
however, is gone virtually the instant you release the key. While you 
can sustain it indefinitely by simply holding down the key, its release 
time is extremely quick. 

The following diagram (Figure 4.1.3) expresses the operation of 
the four ADSR controls graphically. Since the main job of the envel- 
ope generators is to control the attack and decay characteristics of 
sounds (as we shall see, the envelope generators also have some less 
important applications), it is convenient to describe the operation of 
the ADSR controls by relating them to the sounds. 


Figure 4.1.3. ADSR envelope. 

Experiment 2: Settings of the ADSR Controls 

Try the settings of the ADSR, using the patch shown in Figure 4.1.4. 
See if you agree that in terms of attack, decay, sustain, and release, 
these envelope settings are representative of the characteristic sound of 
each instrument. Experiment with the basic settings to see if you can 
refine them to produce an even more realistic envelope, based upon 
what sounds right to you. 

Experiment 3: AR Control of the VCF and VCA 

The AR generator functions much the same as the ADSR; the pri- 
mary difference is that it has fewer controls and creates a less elaborate 
control voltage, one which permits you to control only the attack and 
the release of a sound. For this reason, the AR generator is somewhat 
less flexible than the ADSR, and is generally used for different pur- 
poses. With the AR alone you can create a perfectly acceptable bowed 
string envelope, or you can create the same organ effect that was 
created using the ADSR in Experiment 1. Set the Odyssey up according 
to the patch in Figure 4.1.4, except to change the position of the 
ADSR/AR switch directly under the HPF box to the AR position 
(down). Now the VCF will be controlled by the AR rather than the 
ADSR. Try the following envelopes. 


J a < J— 

5 S V 



=# 2 



^ CD 


,oa> o 



*§ e r 

w W <r 


■ ■ 


— ^ 

5 SS 

-J I 1_ 

=*HI mil 


4.J 1 1 L. 


. .. 

I * • 


f 2 ti 



t t 

=•1 scpk 


1 1 

I ■ 






Experiment 4: ADSR Control of Pulse Width 

As you discovered in earlier experiments, as you change the width 
of the pulse waveforms generated by both of the voltage-controlled 
oscillators, you change the sound of the waveform. You have already 
used the Pulse-Width slider, controlled manually, to demonstrate 
this fact. That sort of control, however, once the width is set, still 
results in a static waveform having a steady unchanging tone. To 
prove this, set up the patch in Figure 4.1.5, next page. You may move 
the pulse-width (blue) "Width" slider of the VCO-1 to any position, 
but as soon as you stop moving it, the sound stops changing. 

Why does this matter? It matters because if you looked at the 
waveforms of many conventional musical instruments on an oscillo- 
scope, you would see that waveforms are not constant but are always 
changing as the sound sustains and then dies away. Such waveforms 
are dynamic waveforms; their timbre is constantly changing, even 
though these changes may be slight. By ADSR-modulating (changing) 
the pulse width of the oscillator, you can create a dynamic waveform, 
changing in harmonic content according to the shape of the ADSR 
envelope. To demonstrate, simply raise the pink attenuator marked 
"Mod" under the Pulse Width bracket on VCO-1. Listen as the sound 
of the waveform changes when you press a key. The change is being 
created by the ADSR control voltage which is, for convenience, being 
triggered by the Keyboard. 

Control Voltage 





Figure 4.1.6. Block diagram 
of Figure 4.1.5. 

You'll find dynamic waveforms extremely useful in recreating re- 
alistic instrumental effects such as a guitar, bass, or piano. Try the 
Electric Bass patch shown in Figure 4.1.7. Vary the pink Mod atten- 
uator on VCO-1 to hear the effect that is being created by the ADSR 
modulation of the pulse width. You may find that you would prefer 
a slightly different bass effect than that shown in the patch given here. 



o SC 

41 o fS 


s| ■ 

d UJ > , 


In ; 


13 j It 
* O 



1 1 1 r 


f. 1 





-> S ■■■■ 




t i 


o z o 

* P g 



















Experiment 5: ADSR-Cont rolled Frequency Modulation 

The final experiment in this lesson utilizes the ADSR voltage to 
control the pitch of VCO-1. You'll find that this control voltage, 
when combined with the prewired control voltage from the keyboard, 
creates an unusual effect — much like an instrumentalist or vocalist 
"sliding" up to every pitch being played or sung. 

The key to this patch is the setting of the ADSR slider controls. 
These will determine what effect the ADSR will have upon the pitch 
of VCO-1, and, with a bit of experimentation, you'll find that many 
ADSR settings do not produce particularly useful pitch effects. To 
guide your experimentation, you will find: 

1. The attack slider determines the "rise time" that will elapse as 
the oscillator goes from the basic frequency (pitch) that you have set 
manually, to the pitch that the ADSR will ultimately "hold" until 
you release the key on the keyboard. 

2. The sustain slider determines the pitch the ADSR will hold for 

3. The release slider determines how fast the pitch will fall from 
the ADSR-controlled second pitch to the original pitch set manually. 

While the control settings will vary slightly from one Odyssey to 
another, the patch shown in Figure 4.1.8 is programmed to do the 

1. The ADSR modulation of the VCO-1 frequency will have a rela- 
tively fast "rise time," as determined by the attack slider. 

2. The yellow attenuator on VCO-1 should be fine-tuned to limit 
the pitch rise to exactly one octave. Adjust the controls on your 
Odyssey accordingly. 

3. The release time will be relatively long, during which you'll hear 
the pitch of VCO-1 fall back to the original octave determined by 
the manual setting of the frequency controls. 

From the Library of 

Sohn 2". 3Hc®aniei 

Lesson 2: The Low- Frequency Oscillator 

The Low-Frequency Oscillator (LFO) belongs to the group of com- 
ponents known as controllers. The output of the LFO is never used 
as an audio signal, but rather is used exclusively to control other func- 
tions of the Odyssey. In addition to the pink LFO Freq slider shown 
in Figure 4.2.1, locate all other LFO-related controls as illustrated. 













Experiment 1: Vibrato. 

Make sure that the LFO switch under the pink FM slider of VCO-1 
is in the sine wave ( Ky^^S ) position; then, gradually raise the pink 
attenuator above that switch — you'll hear a vibrato. As the slider is 
moved to higher positions, the slight pitch change brought about by 
modulating the frequency of VCO-1 with the LFO becomes greater. 
Lowering the slider has the opposite effect. 

The vibrato effect you heard was created by the control voltage 
supplied by the LFO. In this case, the control voltage was in the form 
of a sine wave. At this point, it would be a good idea to study Figure 
4.2.2. Notice that the sine wave is balanced on either side of the 
zero-volts line, while the square wave produced by the low-frequency 
oscillator is "positive-going" only. 

Zero Volts 




Figure 4.2.2. Comparison between sine wave and square wave. 

It is important to understand this if you are to use the LFO most 
effectively: when controlling an oscillator, a positive-going voltage 
can only drive the pitch up (raise the frequency), while the sine wave 
which goes positive and negative will alternately both raise and lower 
the frequency relative to the oscillator's initial setting. Thus, to create 
a vibrato, which is a cyclical one-pitch deviation (alternately above, 
and then below, the original frequency) you must have a control 
voltage which goes both positive and negative, as shown above in 
Figure 4.2.2. The sine wave answers this need by providing a smoothly 
changing voltage which, when applied to the control input of either 
oscillator, creates a perfect vibrato. 

As you slowly raise the pink attenuator, you can increase the 
vibrato "depth" to the point where the LFO is driving the oscillator 
both up and down more than an octave in each direction. Before 
leaving Experiment 1, try different attenuator positions in order to 
"tune" the deviation from the oscillator's basic pitch. While a normal 
vibrato deviation is less than one semitone and usually repeats at the 
rate of 4 Hz to 8 Hz (LFO Freq setting) many interesting new effects 
are possible with wider deviations and faster and slower rates. Try 
varying the pink LFO Freq setting to get faster and slower vibrato 


Experiment 2: Trill 

Set the controls exactly as you did at the beginning of the previous 
experiment (Figure 4.2.1), with one exception: move the two-position 
patch switch on VCO-1 from LFO sine down tc LFO square 
( |~~1 I 1 D- Raise the frequency-modulation slider as you did before. 
This time you will hear a trill. Simply stated, a trill is the alternation 
of two pitches. If you wish to increase or decrease the rate of the 
trill, use the pink LFO Freq slider; to increase or decrease the interval 
(distance) between the two notes of the trill, adjust the pink control- 
voltage input attenuator. Musical intervals from a half step or less on 
up to well beyond an octave are possible. 

The manner in which the LFO is controlling the oscillator is roughly 
the same as in Experiment 1. The only difference is that this time, 
since the square wave is positive-going only, the pitch goes up to the 
level determined by the LFO and then simply falls back to the pitch 
originally set. Listen to prove this to yourself. One of the two pitches 
in the trill will always be the same pitch you originally set using the 
coarse- and/or fine-tuning sliders of the oscillator. 

You should also have noticed that the audible effect of the LFO 
on the oscillator, in both this experiment and the previous one, con- 
forms to the visual representation of the shape of each control wave- 
form used. When the sine wave was employed, the pitch of the 
oscillator gradually rose and fell in a smoothly changing series of 
frequencies; when the square wave was employed, the pitch rose 
immediately from the original pitch, held its level, and then fell 
suddenly back to the first level — just as the shape of the square 
wave rises, flattens, and falls. 

Experiment 3: Voltage-Controlled Pulse Width Using the LFO 

Set the Odyssey controls as shown in Figure 4.2.3. When the Mod 
slider is raised, it now becomes possible to produce a voltage-controlled 
pulse-width change, the rate of which is controlled by the LFO Freq 
slider. While you may find ADSR control of the pulse width (as per- 
formed in Experiment 3 of the previous lesson) more useful, experi- 
ment with the use of a slow sine wave to create a continuously chang- 
ing timbre of the oscillator as you're playing. Some interesting effects 
are possible. 

As with all voltage-controlled functions, you are again simply apply- 
ing a given voltage — determined by the position of the control input 
attenuator — to a voltage-sensitive function. 












< o W| 




o ri cc 

2 ° K 

i 1 — i r- 




A tkm 




k 2 C 


I i si 

& i 

>» : ; v 

I 1 


O J (T 


=01 ^mi^ 

111 col 

As you spend more and more time with the synthesizer, try to 
conceptualize in this manner. Work at visualizing what is happening, 
and why. You will find that after only a short time you will develop 
a "library of mental pictures" that will greatly aid your control and 
understanding of the synthesizer. 

Experiment 4: Controlling the VCF with the LFO 

You'll recall that in Section 3, Lesson 1, Experiment 5, you created 
a tremolo effect by using the pink LFO slider to control the VCF. 
This rapid opening and closing of the filter creates the tremolo effect. 
Try the patch shown in Figure 4.2.4. 

Experiment with different settings of the black VCF Resonance 
control and the yellow S/H-LFO attenuator. As you have 

probably guessed, you are again simply using the LFO to generate a 
control voltage, which, when applied to the control input of the 
VCF, opens and closes the filter at the rate determined by the 
LFO Freq slider. 

VCO- 1 







Figure 4.2.5. Block diagram of Figure 4.2.4. 


Lesson 3: Sample and Hold 

The Sample and Hold circuit of the Odyssey produces stepped out- 
put voltages by systematically sampling portions of the waveforms(s) 
which are routed to its signal input. The voltages which result from 
the action of the Sample and Hold circuit can then be used to control 
VCO-1, VCO-2, and/or the VCF. As you've certainly found from the 
brief experiences with the sample-and-hold function you've already 
had in Part II, some extremely interesting audio effects can be created 
by utilizing the Sample and Hold output voltages to control the 
oscillator(s). In this lesson, you will discover how these control volt- 
ages are created. 

As its name implies, the Sample and Hold circuit does two things: 
(1) it samples, at the command of the LFO or the keyboard, the 
waveform that is being fed into the S/H Mixer (Figure 4.3.1); and (2) 
after the command to sample is past, the Sample and Hold circuit will 
hold the voltage level of the last sample taken until another sample 
command is presented to the circuit. The series of control voltages 
thereby produced can then be routed to the VCOs or to the VCF for 
control purposes. Most often, the voltages produced by the Sample 
and Hold are used to control the pitch of the oscillator(s) and/or the 
filter in oscillation. 





I 1 I Sample command pulses 
Figure 4.3.1. Diagram of the S/H function. 

Now that you have an overview of this circuit's function, let's ex- 
amine a visual representation of precisely how its control voltages 
are produced (Figure 4.3.2). Study the left-hand diagram for a mo- 
ment. You'll notice that the output voltage (shown at the bottom) 
assumes the same voltage as the input signal at the precise instant 
the input signal was sampled. Moreover, you'll see that while the 
voltage of the input signal continues to rise, the voltage output of 
the Sample and Hold maintains the level of the previous sample until 


the next sample is taken. You'll see that the same holds true for the 
illustration on the right; the only difference is that a random, aperiodic 
waveform (noise) is being sampled. As you might expect, sampling 
the noise wave produces output voltages which are random in nature, 
while sampling the periodic waveforms of VCO-1 and VCO-2 will 
produce repetitive patterns that can be controlled by the frequency 
controls of the oscillator providing the input signal. 




I I I I I I I I I | 


1 1 I I ■ I 1 1 I I 

i I I I I I I I 


• I I I I I 
J I — h 

Musical effect. 


i i i i i i i i 


I I I I I I I I 


Musical effect. 

;/nrr JJ Jr 

Figure 4.3.2. Diagrams of S/H operation. 

Experiment 1: Sampling a Low-Frequency Sawtooth to Produce a 
Series of Descending Pitches 

For this experiment, set up the controls as shown in Figure 4.3.3. 
Notice that the waveform being fed into the Sample and Hold mixer 
is a sawtooth wave from VCO-1. This is the waveform that will be 
sampled to produce the control voltages. Note also that the two- 
position switch on VCO-1 (upper right of panel, to the right of the 
blue frequency sliders) is down, in the low-frequency position. 

The "fall" of this waveform will be slow enough to permit the 
LFO Freq setting you are using enough time to sample several times 
on the negative-going ramp of the sawtooth. This series of falling 
voltages is then routed to VCO-2 and applied to VCO-2 as control 
voltages, producing the descending series of pitches you hear. 


VCO - 2 







VCO -2 


Figure 4.3.4. Block diagram of Figure 4.3.3. 

Before going on to Experiment 2, be sure to try various positions 
of the coarse-tuning slider of VCO-1. You'll find that each position 
produces a different pattern of pitches. As you change the frequency 
of the waveform being sampled, you are changing the points at which 
the Sample and Hold circuit takes its samples (Figure 4.3.5). 

-1 Second- 

1 1 • r 
1 1 1 1 1 1 r 

! ! ! ] J 1 1 1 

SAMPLE COMMAND , i ! ! ! ! 

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 
> 1 1 1 1 1 i ■ 1 1 1 1 1 1 1 1 
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 
1 1 1 1 1 1 1 1 1 1 1 1 1 1 i f 


MM p-U I'M 

I I 


0-1 Second— ^ 

iNJi i 

i n r i ii i 

mm i i i i i j j 

sample' command 

kkik Liiilli-Lixi 

i i i 
i i i 

l M M M 
l I I I I l I 

I l I i i i i I I I I 


Figure 4.3.5. Creating different patterns with the S/H. 





5£T 1 I 1- 



< c S! 


O cc 


=> — |iiH=ig 
*H! mis 





I c 

2 g J_ 
E5 t 



S d £ 

513 • 




E 5 CtZ 

I § lit 

t ♦ 



111 mi 



Experiment 2: Sampling White Noise to Produce a Series of Random 

Having now determined in Experiment 1 that repetitive patterns 
can be created by sampling a periodic waveform, the patch shown in 
Figure 4.3.6 will produce a series of random pitch patterns by sam- 
pling the white noise supplied by the noise generator. (If you would 
like a visual review of exactly what is happening in terms of sampling 
the noise wave, review Figure 4.3.2 presented earlier in this lesson.) 

Note that while the series of pitches will still be random, you can 
limit the over-all range within which the pitches fall by moving the 
white noise input attenuator on the S/H mixer down — try it about 
half-way. Experiment with different positions of this attenuator to 
control the range of the sample. (You will also find that a similar 
limiting effect can be produced by simply attenuating the yellow 
control-voltage input to the oscillator.) 








Figure 4.3.5. Block diagram of Figure 4.3.6. 



v r 

i > 

L 1 ! 


t *T 1 i r 


m UJ CC 

3 d H 


• H»<mK 

i « £ 

si 1 

O « 

^=*H* ms$ 
==# — |laa=BS«i 

• HI mil 

■ L_ 




r j , 


2 i ?i£ 


j a j 
O z 5 

* PS 




1 o 



• (6 








Experiment 3: Sampling a Mixed Wave to Control the Pitch of the VCF 
in Oscillation 

When the filter is put into oscillation, the Sample and Hold circuit 
can be used to control the pitch of the audio sine wave produced by 
the filter, just as you can control the pitch of the audio waveforms 
produced by the voltage-controlled oscillators. Set up the patch shown 
in Figure 4.3.7. Notice that this time you are actually sampling a 
mixed waveform; the input from the VCO-1 is providing a low- 
frequency square wave, while the input from VCO-2 is providing a 
square wave that is actually in the audio range. 

You should understand that you can sample audio frequency 
waves just as you would sample any voltage. To prove this, lower 
the white noise/VCO-2 input attenuator and leave the blue input 
attenuator from VCO-1 open (up). Move the two-position switch 
below the blue VCO-1 input attenuator up to the sawtooth position. 
Listen for a moment as the Sample and Hold circuit samples the low- 
frequency sawtooth wave. You'll probably hear patterns similar to 
those you created in Experiment 1. Then move the two-position 
Audio KYBD switch at the top of VCO-1 up, to the audio frequency 
range; listen to the resulting pattern of pitches. Again, you'll find 
that by varying the frequency of the wave being sampled (VCO-1), 
you will be able to vary the pattern of the pitches you hear. You can 
also vary this pattern by changing the rate of the sample, controlled 
by the LFO Freq slider. Experiment with each of these two methods 
of changing the audio pattern. 





_ Trigger 




-TUT, r 

VCO - 1 

Figure 4.3.8. Block diagram of Figure 4.3.7. 


Odyssey keyboard. 

Lesson 4: The Keyboard Controller and Related Controls 

As you have undoubtedly realized by this time, when you are using 
the keyboard, you are simply using another method of voltage control. 
The keyboard of the Odyssey provides an interesting combination of 
voltages, each having its own particular function: (1) a gate signal, 
(2) a trigger signal, and (3) a control voltage. These voltages are inter- 
nally routed to the functions they are most often used to control. 

The function of the gate voltage is to indicate that at least one key 
is depressed. The gate voltage is an on/off type of signal used to acti- 
vate the envelope generators. At the precise instant that any key is 
depressed, a second signal is generated — a trigger voltage. The func- 
tion of the trigger voltage is to indicate the exact instant at which 
any key is depressed. This trigger output appears every time a key is 
depressed, regardless of how many keys are already being held down. 
The trigger signal is also used to start the envelope generators, and 
to trigger the Sample and Hold. 

When any key is depressed, the keyboard control voltage assumes 
some value, depending upon which key is depressed and upon the 
settings of the Transpose and Pitch Bend controls. The function of 
the keyboard control voltage is to provide a control over the fre- 
quency (pitch) of the oscillators, and over the pitch of the filter when 
in oscillation. 

These signals principally serve to provide the sort of response that 
we have come to expect from a keyboard through our experience 
with pianos, organs, and accordions. These voltages are prewired, and 
precalibrated to provide this convenience. When you play the lowest 
C on the keyboard, followed by the C one octave above, the proper 
control voltage is supplied to the oscillators to produce the pitch 
change that is expected. It is important to understand each of the 
three voltages — gate, trigger, and control — and how they perform 
their functions. 


Let us now move on to the variable keyboard controls: Portamento, 
Pitch Bend, the Transpose switch, the Odyssey's two foot-pedal con- 
trollers, and the KYBD CV input attenuator to the VCF. 

Experiment 1: Portamento 

Before beginning this experiment, be sure that the Portamento 
foot switch is not plugged into the Odyssey. 

The Portamento control, when the slider is raised, introduces a 
variable lag, or slide, into the control-voltage output that determines 
the frequency of the oscillators. Its effect is to prevent the control- 
voltage output from responding instantly as different keys are de- 
pressed. This means that the VCO being controlled by the keyboard 
will not move sharply from one pitch to another, but instead will 
"slide" from pitch to pitch. Figure 4.4.1 shows graphically how this 


J Lr ~kr f J lfH - 



Figure 4.4.1. Diagram of effect of Portamento control. 


Set up the patch shown in Figure 4.4.3 and play several notes on 
the keyboard. Play a familiar tune. Then raise the Portamento slider 
and play the same series of notes again. Try different degrees of porta- 
mento. You'll find that while the effect is certainly interesting, the 
movement from pitch to pitch is so slow as it reaches its maximum 
that it is of less use musically that the lesser degrees of portamento. 






- - MAX 









Figure 4.4.2. Portamento control positions. 



I ij J 1 1 L, 




i w rn 
5 < 5 

a d a 
< o £ 

5 5 = 

u, O 1 

5 ii 



o!2 J 1 jafe — I l~. 

<* 1 » 1 — A 


> W 1 



pS — i — r—i — W | 
| 1 






- 4 ; ; i iH 

-<::;: »—i 

I C T 1 jj 


1 1 f ; +■ 

2 I 3 ~ 


> o 1 



Q nr 

i ml 

ill mi 


Experiment 2: The Transpose Switch 

The Transpose sv/itch permits you to extend the range of the 
Odyssey keyboard two octaves in either direction, providing a total 
range of seven octaves. It is simply another voltage controller. Using 
the patch shown in Figure 4.4.4, try the Transpose switch in each of 
its three positions. You'll find that it does make a marked difference 
in the sound of the same patch, depending upon which octave you 
are in. 

If the frequency you are using is a relatively low one to begin 
with, you may find that by transposing down two octaves the pitch 
becomes a series of subaudio "clicks;" if the frequency with which you 
began was a relatively high one, by transposing up two octaves you 
may go above the range of human hearing. For this reason, most of 
the patches in this guide are based upon the basic frequency settings 
of the oscillators as shown in Figure 4.4.4. This will provide the most 
flexibility in your own patches as well. 





pitch BEND 









Figure 4.4.5. Transpose switch musical effect. 





u. O 1 

> J 5<= 


I o. — 


Uj s 

< o§ 
5 £ ^ 

5 £ 


*Hi mil 

S . i 

£ t / >_* \ 

1. j » ' g 
— 1 

x § 

—ill !' » 1 

— < i ; ;-*H 

>£LJLJS § 


X 5 
JL t. 

« O 3 „ 


-0 j t 4nr^ fti| 


p i i r 






1 //^nX 5 


Experiment 3: Pitch Bend 

Below the Portamento slider, you will find a knob marked "Pitch 
Bend." You'll find this control useful in at least two ways: first, it 
will permit you to tune easily to other instruments after you have 
tuned the two oscillators with each other. You'll find this a conven- 
ience that will save time, eliminating the need to bring one oscillator 
into tune with whatever instrument(s) you are going to play with in 
ensemble, and then tuning the second oscillator with the first. 

The second primary use for this control is as the name would imply: 
you can use it to "bend" pitches, in order to recreate more realistically 
the effect guitars, other stringed instruments, and even some wind 
instruments can create. Naturally, you can use it to create imaginative, 
unique effects of your own as well — effects that need not be imita- 
tions of any existing instrumental sound. 

When recreating the effect of the pitch bend of more traditional 
instruments, however, it would be wise to limit the pitch deviation 
to about one half-step. This is the most common effect employed, 
even by rock guitarists. Set up the patch shown in Figure 4.4.6, with 
the Pitch Bend knob set at twelve o'clock, as shown. You will find 
that it takes a little practice to make the pitch deviation smoothly 
and return to the original pitch. Some performers prefer to set the 
Pitch Bend knob all the way counterclockwise so that, after bending 
the pitch upward, by simply rotating the knob fully counterclockwise 
again the pitch returns to the original oscillator frequency. The one 
disadvantage to this system, of course, is that it will not permit you 
to tune the Odyssey to other instruments, unless you tune only in 
an upward direction within the limits of the Pitch Bend knob's 
control voltage. 

Figure 4.4.7. Block diagram of Figure 4.4.6. 



A li li 

1 Voltage 








Experiment 4: Using the Keyboard Control Voltage to Create Keyboard 
Intervals of Less Than One Half -Step 

You will find that larger, studio-model synthesizers often have a 
provision by which you can alter the keyboard control voltage so 
the total interval of, say, a four-octave keyboard might be less than 
one octave in terms of actual pitch. The musical result would be, of 
course, an "octave" with as many microtones as there were keys on 
the keyboard (in the case of a four-octave keyboard, forty-nine). You 
can approximate this effect on the Odyssey by setting up the patch 
shown in Figure 4.4.8. (This effect was previously observed in our 
discussion of the VCF.) 

Having set the controls as shown, raise the black KYBD CV input 
attenuator all the way. When you play a few notes on the keyboard, 
you will discover that the filter (in oscillation, to produce the pitch 
you are hearing) is responding normally in terms of the musical 
response you expect to hear from a keyboard. Now lower the 
KYBD CV slider a little and again play a few keys on the keyboard. 
Try a tune you know well. You will find that the keyboard is now 
responding with intervals of less distance than you would expect. 
Lower the slider again, and repeat this exercise. You will soon find 
that you can create all sorts of microtonal patterns, depending upon 
the degree to which you attenuate the control voltage of the keyboard 
as it is applied to the filter. 







Figure 4.4.9. Block diagram of Figure 4.4.8. 

As an interesting conclusion to this experiment, see if you can tune 
the keyboard to respond with quarter-tone intervals where there 
would normally be half-step intervals. This will mean that by playing 
the lowest note on the keyboard and following it with the same note 
two octaves higher, you should hear an interval of only one octave. 
You may want to return the black slider to the highest position in 
order to use the keyboard to fix the octave interval in your mind 
before tuning with the slider to create the quarter-tone intervals. 







2 ° — 


> o 

dp— |g rags 

I ml! 




— e 



I 'A i 





l CX3|< 

111 mi 

■1 ll 

: LU 


Experiment 5: The Foot-Pedal Controllers 

Two foot pedals are supplied with the Odyssey: the first is a small, 
square pedal that acts like an on/off switch for the Portamento control 
slider. When this foot switch is plugged into the appropriate jack on 
the rear of the Odyssey, the portamento function (described in Lesson 
4) is disconnected except when you step on the pedal. This feature 
allows you to preset a certain amount of portamento on the front 
panel control and then to introduce this portamento by stepping on 
the switch. 


The larger pedal, which operates somewhat like the accelerator on 
an automobile, can also be plugged into the back of the Odyssey, 
permitting you to control with your foot certain functions shown on 
the control panel. When this pedal is plugged in, it may be used to 
control the pitch of VCO-2 and/or the opening and closing of the 
filter (VCF). Quite a number of interesting effects can be created by 
using the pedal to control VCO-2, including a totally new sound in 
music performance — phase-synchronized oscillators. You'll also find 
that by using the pedal to control the VCF, you can very easily create 
an excellent wah-wah pedal effect. Set up the patch shown in Figure 
4. 4. 10. This patch will provide the basis for experimentation with both 
the foot switch and the variable pedal. 

First, play a series of notes on the keyboard, using the foot switch 
to introduce portamento only when you want it. If you have a fairly 
long portamento time set by the control slider, you will find porta- 
mento most effective only when you come to a note that can be 
sustained long enough to permit the pitch to slide up or down to the 
key you are playing. In musical terms, this would mean a half-note 
would be required in the melody, possibly even a whole' note. 

Second, experiment with the effect of the variable pedal controller, 
using it first to control VCO-2, and then the VCF. All that is neces- 
sary to introduce this control voltage to either of these functions is 
to raise the attenuator above the two-position switch on each function 
that is labeled "S/H Mixer or Pedal." When the pedal has been plugged 
into the back of the Odyssey, the S/H Mixer is automatically dis- 
connected from both control inputs and is replaced by the pedal 
control voltage. 


■XX * 




Experiment 6: Polyphonic (Two-Voice) Keyboard Capability 

Until only very recently, keyboard controllers on all synthesizers 
could play only one note at a time. These units were designed to 
synthesize instrumental voices, and logically, since a clarinet or an 
oboe cannot play more than one note at once, the synthesizers were 
designed to function in a similar manner. While this is not taking 
into account the additional technical difficulties in creating a multi- 
voice keyboard, the fact remains that most synthesizers still have 
monophonic keyboards. 

Keyboard performers, however, are accustomed to polyphonic 
(chord) response from their instruments. The Odyssey, therefore, has 
been designed to permit the player to play two independent pitches 
at the same time. This is accomplished by having the VCO-1 play the 
lowest key of any two notes being played, while VCO-2 plays the 
higher key. If you are a keyboard player, you will find this feature 
extremely useful in both composition and live-performance situations. 
For a basic two-voice patch, use the control settings shown in Figure 
4.4.11. Feed both oscillator signals into the audio mixer and tune the 
frequencies to exactly the same pitch (unison). This can be accom- 
plished by holding down the lowest key on the keyboard with one 
hand and tuning with the other. You can also manually open the 
filter and the VCA for the purposes of tuning. After you have tuned, 
play a brief melody or phrase you know well, using the two-voice 
capability. While the patch shown is designed to reproduce the tra- 
ditional response in terms of pitch, as related to the intervals you are 
playing on the keyboard, you should not ignore the possibility of 
different oscillator tunings to create different effects. Try these, for 

1. Tune VCO-1 and VCO-2 to the same pitch, but one octave apart. 
This tuning provides a bright-sounding, useful effect. 

2. Tune VCO-1 and VCO-2 a major third apart (a major third, in 
terms of keyboard distance, is five steps — counting black and white 
keys, and counting the basic pitch as "one" and the key that produces 
the pitch of the major third as "five.") You'll find that this tuning 
produces surprising musical results the first time that you play a two- 
voice line that you know well enough to anticipate what you would 
hear if it were played using traditional tuning. 

3. Tune the two oscillators to other intervals, including a fourth 
apart, a major second apart, and a minor second apart, as well as any 
interval your own curiosity might suggest. 



22 sV 



J s 8 35 

1 Mi 



* si 


• JgazoM 

■ |j A 1 

pS — i — 




t=m — nico! 

f 8 

I • 

O z O 





i § fi 
1 . 1 . 

O j 5 
o z o 

*1 scn|< 

||| col 




At this point you have now completed those parts of this text 
designed to (1) acquaint you with the fundamental terms and princi- 
ples of electronic sound synthesis, and (2) to apply them specifically 
tc the operation of the ARP Odyssey. By this time you should be 
thoroughly familiar with the location and function of all the controls 
of the Odyssey; you should also have firmly in mind the information 
presented in Part L Ideally, that information has been reinforced by 
the experiments performed in Part II, so that you can now move on 
to musical applications without being hindered by having to refer con- 
stantly to these sections to perform even the simplest experiment. 
Naturally, a certain amount of reference to that material is inevitable — 
and is encouraged at points in Part III which involve recalling specific 
patches or exercises performed earlier. On the whole, however, it is 
assumed from this point on that you have the basic knowledge re- 
quired to undertake more sophisticated musical applications. 

Part III presents separate sections on five general topics: timbre, 
melody, harmony, transposition, and setting up an electronic music 
studio. Within each section, however, you will find related informa- 
tion, ranging from a discussion of ear training to a full-length subsec- 
tion on tape music techniques. In the event that you are not taking 
the material sequentially, refer to the Index for page locations of 
specific subjects. 


Section 1: Timbre 

The qualities of a particular sound that enable you to distinguish 
that sound from any other sound can be generally defined as timbre. 
A number of experiments in Part II of this text were designed to use 
various functions of the Odyssey to change the timbre of the sounds 
being produced. Moreover, the Odyssey — unlike a trumpet or a 
violin — is capable of radical modifications of timbre. While a trumpet 
(despite all you might do short of recording it and then modifying 
the recorded signal) will still sound like a trumpet, the Odyssey per- 
mits almost total timbral flexibility. It therefore is ideally suited to 
the systematic explanation of the subject of timbre. 

Timbre is created and/or affected by a number of interacting fac- 
tors. Think for a moment about why a flute sounds different (has a 
different timbre) from a guitar. Your study of Part I of this text will 
provide a number of basic answers. You'll recall that it was established 
that different sounds have differently shaped waveforms. It would 
follow then, that the flute and the guitar obviously produce differ- 
ently shaped waveforms; taking this one step further, based upon in- 
formation also contained in Part I, these instruments produce 
acoustical waveforms containing harmonics differing in number and 
amplitudes. For example, a flute produces an almost pure sinelike 
wave; the guitar produces a waveshape more like the dynamic pulse 
of the Odyssey — containing almost all harmonics, but with an ever- 
changing harmonic content from the time that the string begins to 
vibrate until it ceases vibrating. 

This changing nature of the harmonic content of an instrument's 
characteristic waveform brings up an important point: the harmonic 
content, and therefore the timbre, of waveforms produced by tradi- 
tional instruments is not constant but, as stated earlier, varies consid- 
erably from the moment the sound begins until it dies away. When any 
such instrumental tone is sounded, the build-up of sound from zero 
to maximum initial volume is a complex process termed the "attack 
transient." During this time, pitch, volume, and the spectrum of 
over-tones go through complex changes to which the human ear is 
extremely sensitive and perceptive. For instance, a wind instrument 
usually starts its attack by producing only the fundamental tone, and 
higher harmonics are added as the sound builds up (Figure 1.1). This 
is because the column of air in a wind instrument starts to vibrate in a 
simple harmonic vibratory mode, but breaks up into a more complex 
set of vibratory modes as time progresses. 

We have established, therefore, that attack and decay, as well as 
the basic waveshapes or harmonics of any sound contribute measur- 


AMPLITUDE f*" ° ne Cycle 


Figure 1.1. Attack transient pattern. 

ably to the timbre. While these may be the most important, timbre is 
also affected - or more properly, our perception of timbre is affec- 
ted - by such things as volume, the pitch (frequency), the presence 
of a vibrato or tremolo, and even the duration of a sound. Other 
factors may occur to you; if so, test them against the basic definition 
that timbre is any quality which makes a sound different or distinc- 
tive. Be sure, however, to also take into account overlapping ways of 
saying the same thing. 

For example, it might be argued that the timbre of a large flat-top 
acoustic guitar is different from that of a solid-body electric, even if 
the acoustic guitar is also amplified. Therefore, it would seem to follow 
that the difference in the construction and materials of the two guitar 
bodies is a factor affecting timbre. To a point, this is correct; however, 
carry it one step further by reducing the timbre difference to the low- 
est common denominator of sound synthesis - the waveform. The 
reason that the bodies of the guitars have an effect upon the timbre 
is because they each act as a filter/resonator, absorbing certain fre- 
quencies while emphasizing others. Thus, the net result is two differ- 
ent waveshapes, just as it would be if you began with the same initial 
sound source on the Odyssey but processed it two different ways 
with the VCF and the HPF. 

It is important to remember, then, that while considering factors 
affecting timbre, your key to real understanding is to relate tangible 
physical factors (such as the guitar bodies) back to the basic com- 
ponent parts of any sound. By doing so, you will not only enjoy a 
greater understanding of the world of timbres around you, but you 
will also have a significant head start toward reconstructing or 
synthesizing those sounds you've heard. 

The experiments in this section are designed to further your know- 
ledge of timbre by permitting you to change and carefully control a 
number of those factors that contribute to the distinctiveness of 
several sounds. If you have access to a tape recorder and can get 


together a number of students who play different instruments, you'll 
find the first experiment provides a dramatic illustration of how 
timbre is affected by the attack and decay characteristics of a sound. 
The same experiment, of course, can be performed utilizing the ADSR 
envelope generator of the Odyssey. Your understanding will benefit, 
however, if you actually hear what happens to the sound (timbre) of 
familiar, conventional instruments when robbed of their distinctive 
attack and decay. 

Experiment 1 

On a day that has been agreed upon in advance, have several stu- 
dents bring their instruments to class. If possible, try to have at least 
six to ten different instruments represented. Perform the following 
sequence with each different instrument. 

1. Turn on the recorder and set it to record. A tape speed of l l A 
ips (inches per second) is best. 

2. When the tape is rolling, signal a student to begin a single tone 
and hold it until the signal is given to stop. Each tone should be sus- 
tained at least five seconds. Wind instruments will have no difficulty 
in doing this, nor should bowed strings. Other instruments, such as a 
guitar, a piano, a triangle, or a xylophone, should be struck once and 
then simply allowed to ring. Have each student play the same note 
(C,, for example) but in the mid-range of his own instrument. 

3. When all instruments have been recorded, edit the tape so that 
the total attack and decay of each instrument has been cut out of the 
tape and the remaining parts - the two or three seconds of continuous 
tone - spliced back into the reel with brief silences between each 
example. (See the section on Tape Editing for techniques.) 

4. Now play the tape and listen for the differences in the timbre of 
each instrument. In some instances the change is so radical that the 
instrument will be difficult to identify; in almost every case, it will 
be agreed that the attack and decay characteristics contribute measur- 
ably to the distinctive timbre of each instrument. 

Two enjoyable variations of Experiment 1 can also be performed. 
The following variations are given for those classes of students wishing 
to pursue further this aspect of our study of timbre. 

Experiment 2 

Perform the four steps outlined in Experiment 1. This time, how- 
ever, have the students play a note as close in pitch as possible to that 
being played by all other instruments. Try middle C or, if you have 


the proper pitch pipe or tuning fork, A-440. After editing, you'll 
probably find the instrumental timbres even more difficult to identi- 
fy — particularly if you change the order or sequence of instruments 
from the order in which they were recorded. 

Experiment 3 

Save the pieces of tape edited out of the main reel — those pieces 
having the attack and decay recorded upon them. Splice all of the 
attack segments together with brief silences between. Are these seg- 
ments easier to identify then the central portions retained on the 
primary reel? You may also want to try the decay segments only. 
Other combinations, such as splicing the attack of one instrument to 
the decay of another, will occur to you as you experiment. These, 
however, are beyond the immediate scope of our discussion here as to 
how the attack and decay characteristics contribute to the distinctive 
timbre of any instrument. 

Relating it all to Synthesis 

Throughout this text, most subjects are approached from two di- 
rections: that is, starting from a central concept, such as timbre, it is 
intended that (1) your understanding of traditional musical concepts 
and/or phenomena will be expanded, and (2) at the same time you 
will also apply this knowledge as it relates to the Odyssey — thus 
expanding your ability to recreate sounds you've heard and to syn- 
thesize totally new sounds with an ever-growing degree of ease and 

Relating the attack and decay of instrumental sounds to the 
Odyssey, we find that in order to achieve the expressive qualities of a 
traditional wind instrument, it is necessary that an electronic musical 
instrument be able to "process" waves to build up in a similar way, 
from a simple harmonic mode to a complex wave. In order to ac- 
complish this, electronic musical instrument designers turned to a 
device known as a "voltage-controlled low-pass filter." This type of 
filter has a variable frequency response, in that its operation can be 
varied to allow more or fewer high-frequency waves to pass through 
relative to the low frequencies. Its operation can be shown by con- 
sidering how a complex wave with a large number of harmonic over- 
tones is modified, as shown in Figure 1.2. 

In the actual operation of the Odyssey, the control of the fre- 
quency response is automatically achieved by application of the con- 
trol voltage generated by the envelope generator(s) to the voltage- 
controlled filter (VCF), whose frequency response is a function of 
the control voltage. 






Minimum Middle Maximum (Wide Open) 
Frequency Frequency Frequency 
Response Response Response 


r \y\. r^f\j run 








Figure 1.2. Diagrams of filter frequency response. 

The waveform resulting from this processing by the VCF has the 
attack and decay characteristics preprogrammed by the ADSR or AR 
generators. Therefore, in Figure 1.3, you'll see that the harmonic 
content of the square wave begins (during the attack time) the gradual 
build-up spoken of earlier. Similarly, during the release time, the 
harmonics die away in accordance with the preprogrammed decay or 
release time. Visually, then, you now understand those portions of 
the instrumental sounds that were edited out of the tape in 
Experiment 1. 





Lgf. TIME 



DECAY (Release) 


Figure 1.3. Creating an attack and release with the VCF and Envelope Generator. 
Experiment 4 

This experiment will provide an opportunity to "edit" with the 
ADSR generator of the Odyssey. In this instance, however, you'll 
have the ability to control totally each of the four parameters — 
attack, initial decay, sustain, and release — that contribute to the 
timbre of the sound you are creating. 

You may begin this experiment with the clarinet patch shown in 
Figure 1.4. You should, however, try a number of the instrumental 


patches that appear at the end of this section. In each instance, you'll 
find that as you modify the settings of the ADSR, the realism of the 
patch is very much affected. 

If you want to simulate the "chopping" of all attack and decay 
characteristics, as was done by editing the tape in Experiment 1, set 
the ADSR as shown in Figure 1.5. 

Figure 1.5. Control settings for attack and decay experiments. 

Experiment 5 

To demonstrate easily the effect that the straight harmonic content 
of a wave has upon timbre, set up the patch shown in Figure 1.6. Now, 
without the interacting factors of attack and decay, you can dem- 
onstrate that the basic harmonic content itself is a determining factor 
in the timbre of any sound. To do so, gradually raise the pulse-width 
slider on VCO-1 from 50% to 10%. You'll hear both the number and 
amplitude of the harmonics increase as you move toward 10%, pro- 
ducing a brighter, more nasal sound. For a visual reference as to what 
is happening, refer to the following diagram, (Figure 1.7) reproduced 
from Part I. 

Note that when the pulse-width slider is in the 50% position, the 
square wave produced contains only the odd-numbered harmonics; as 
the wave shape changes to a narrow pulse, the even-numbered harmon- 
ics are added and the amplitude of the higher harmonics increases 

Before going on to Experiment 6, you should also listen to the saw- 
tooth wave from VCO-1 and the sine wave produced by the VCF in 
self-oscillation. If you need to review how the patch for each of 
these waveforms is created, refer to the appropriate sections of Part II. 




1 1 ■ 








O £ u 
> o 

=^=*— II mil 

* 2 «< 



t ? 

| sv i — r 

1 1 £§ 

♦ t 


=!=# o 



*** * 



g a 




a C 

32 gV 

L 1 ! 



u, ui ee 

u j a 

< o it 


* o < 



>— Re- 

S 3 tc 


• 1 mil 


K I 

S 8 .| 

! 6 3 





ui 3 O 









of the 




— e- 

J\ L 





F. 3rd 5th 7th ... 

F. 2345678 9. 

F. 2 34 567 89.. 

Figure 1.7. Harmonic series for different waveforms. 

Experiment 6 

Set up the electric bass patch shown in Figure 3.1.13. You 
are now ADSR-modulating the pulse width (harmonic content) of 
VCO-1. To demonstrate the difference in timbre that this is creating, 
simply lower the input attenuator that is permitting the ADSR to 
affect pulse width. You should also try various settings of the input 
attenuator, in order to achieve maximum realism to your ear. 

Experiment 7 

Earlier in this section, we stated that the pitch of a sound affects 
the way we perceive timbre. Take the same electric bass patch 
you used in Experiment 6 and transpose the effect up two octaves 
by using the Transpose switch. The result, of course, is totally unlike 
an electric bass; try the same procedure with the tuba patch. If you 
could actually see on an oscilloscope what is happening to the 
waveform, you would see that the basic waveshape has not changed; 
the frequency has simply increased. Technically, then, when there is 
no change in the basic shape of the wave, there can be no change in 


Is ^ 

m it 

53 §S= 

m iu a 
O 3 - 
2 O !t 




* 5 < 

-> a - 




>{S r~ 

SIS " 


t ♦ 

81 9 

• jlmll 


— jivEQ i 














the sound or timbre. What does happen is a twofold effect: first, by 
raising the fundamental two octaves, the harmonics are correspon- 
dingly higher, though in the same relationship to the fundamental. 
Nevertheless, by radically raising the pitch, it is possible that the upper 
harmonics will be above audible range, thus changing your perception 
of the timbre. Conversely, if you lower any effect so that the funda- 
mental is below the lowest frequency humans perceive as pitch, you 
will again drastically change the perceived timbre, without changing 
the waveshape. This is one reason why two instruments producing the 
same basic waveshape in different frequency ranges may sound quite 

The foregoing hints at the second point: we associate certain pitch 
(frequency) ranges with certain instruments through our own musical 
experience. Therefore, an electric bass transposed up beyond its char- 
acteristic range is simply inconsistent with what we expect to hear 
and the instrument's traditional musical purpose of providing a rhyth- 
mic and tonal foundation. More on this subject as it relates to orches- 
tration appears later in Part III under "Melody." 

Experiment 8 

The final experiment in this section demonstrates how a vibrato 
or tremolo affects the timbre of instrumental effects. Set up the flute 
patch shown in Figure 1.8. 

First, raise the input attenuator on VCO-1, permitting the LF sine 
wave to create a vibrato. A vibrato, of course, is a cyclic pitch change — 
the control voltage of the LFO actually raising and lowering the fre- 
quency of VCO-1. As you raise the slider further, permitting the 
vibrato to grow wider, the timbre is an effect totally uncharacteristic 
of anything traditional, except perhaps a musical saw or certain types 
of police sirens. Is a timbre change involved? In the strictest sense of 
basic waveform shape, perhaps not; nevertheless, as in the case of the 
"soprano" electric bass, our perception of the timbre is certainly 

Now lower the attenuator on VCO-1 and raise the input attenuator 
for the LFO on the VCF. Again, play a few notes using the flute 
patch. Include a few sustained tones in your playing. Does the tremolo 
effect create a change in timbre? Definitely. What is actually occurring 
is amplitude modulation of the upper harmonics, controlled by open- 
ing and closing the filter with the low-frequency sine wave. Just as 
in the case of the ADSR control of the filter we examined earlier, 
here an actual change in the waveshape does take place as harmonics 
are weakened or totally removed. 


5 2 a '"~ 

I * ■ 


< _ 

I m Pi 



*5£S 1 1- 


O a 
5 S 

• || mil 


o o 

3 gl 

r . 



O rr 

o o 

»1 scq< 

t t 

s g g . 

pq 1 


//ANN I 


! II fe 

m It 

(t UJ » 
«/> C. « 

32 g 



(J js 

X a j 

— r 


I 5 
sir 1 — 


u. O I 


li mis 



O 3 3 



— Ifiuuf!; 


5 n 


if I - 


*, J 

S . B * a 






if 1<£ 


< O t 


O jit 


' il 


> oc 


J ° 5 


6* ; 










9» T- 




^ 1 r 

t t 


O 2 O 
> O W 

d* — 1|1 








2 a /J— 



S P5 


53 < " 
i a j 


ui w 


8 j<=J=f : 

\ I/I 




*HI mil 

t &CjLj I 


5: Z 

s oil/ 

Jj 8 >OZ& §j*| 


I \ he 


>1 SCE)^ 




• < 





/s ! 

go a 


a « P 
5 E 


* 2£ 

o t e 



- «!!■'! »HI mil 

>• * 




— ! 



J ^ v j — r~f* gj 



=#1 scq< 






2 1— 
15 9 ■ , - , , ■ 

Is — — s riJwng? 

» L 




s|s a,i- 



1- 5E 


* o •< 

> o 

: : : : » — Kuk 
; : » :» H8- ™ 

3 S3 

s L 1 1 1 ■ 

* p i 1 r— 

O E 

5 S 


=^=*HI mil 


I J 

| I <==*= 

J 3 



s 1 

? § 

♦ t 


i 1 r 


I Eh 









■/ 5 § S- 

r % 

0 « <<u S 

cm > 

to a « J 1 a 

DC * 2 

3s 2 

t v i i — i i 

=» [lit 




x a. = 

s El 

O <r 

•-2/ — 


=(» — If KZDIS 





i $ If 

f ♦ 

g I i ' 


i mi 




Is - 

OC UJ > , 

v> a. < J— 

90 i|C 

< r 1 . O ' 

m UJ oc 
O -j w 



1 S E 


j i — i— 

Aj — t- 

o ■ 
5 x 


— s= 

-S. — liii-»SS 



^ 1 i 








S3 e 

£ o £ 


■ 2s 

tr * £ 




>H S( ™ 


5 g 
l t 

£ ° 3 

, t f 

ui "J O 






S B 5 

< o = 



3 5 r> 

-13 < 

> — \&W=l 


II mil 


♦ ♦ 

SI 3 « 

5 In 

t I 

I 32 



=#1 s CON 










z0 5 1 sens 


To 5 


H i 

' o < 

» : : ; : »Hi ■ ; 

(3 jl 

j F j 



u. O I 

O ac 


3 - 


2 Si "J_ 


1 1 
, ♦ t 

Q a 


t ) 






Section 2: Melody 

Have you ever asked another person to whistle or hum a tune which 
you were trying to recall? If so, you were asking the person to pro- 
duce the pitches which make up the melody of a song. In a sense, 
melody is a kind of musical identification tag, as evidenced by our 
ability to recognize a large symphonic work by just one short melodic 
theme. Familiar songs are usually quickly remembered with the play- 
ing or singing of a few of the melody notes. You may have heard 
melody referred to as the "tune" of a song — it's just another com- 
mon way of identifying the important aspect of melody. 

The actual effect of a melody upon the listener, however, is depen- 
dent upon several interacting factors, all of which can be classified as 
submelodic component parts. To illustrate this concept, an analogy 
can be made to speaking; we all — know people who — talk — 
in — a — dull — monotone voice just like — this. You may know 
others who talk so fast when they are excited that theirwordsrunto- 
getheralotlikethis. Inflection (the pitch of someone's voice), speed 
(tempo), rhythm (the way they say a phrase and the pause before con- 
tinuing), and the basic tonal characteristic that makes your voice dif- 
ferent from anyone else's (timbre) all play a part in the effect your 
words will have upon your listener. Similarly, in a melody, rhythm 
(the duration of the notes), tempo (speed), the relationship of the 
pitches (tonality), and the timbre (the qualities of sound that make it 
distinctive) interact to create a specific effect for the listener. A musi- 
cian's awareness and control over these submelodic components will 
largely determine whether the melody is uninteresting, or even un- 
pleasant, as opposed to being a melody that has all the potential of 
becoming a part of something wonderfully exciting. 

Let's now look at some specific examples of how the character of 
melody is affected by the control of the musical aspects mentioned 
above. To begin, set the Odyssey controls as shown in Figure 2.1 and 
you'll be ready to examine melody in terms of rhythm. 


Experiment 1: Rhythm and Melody 

Using the patch shown in Figure 2.1, play the melody in Figure 2.2. 
Next, in every measure which contains four quarter-notes (measures 
1, 5, 6, and 7) change every other quarter-note to a half-note, and play 
the melody again. Does the melody sound the same? The suggested 
change is not very great and the melody is probably still recognizable; 
however a radical change of time values in the melody of Figure 2.2 
could eventually make the tune unfamiliar. Try this experiment with 
a friend: see who can make the simplest and most familiar melody 
unrecognizable by changing the rhythm. 

By performing this experiment, you are actually demonstrating 
that melody and rhythm cannot be thought of as being mutually exclu- 
sive musical phenomena. Thus, the two fundamental qualities of musi- 
cal sound — pitch and duration — form the raw material from which 
all melodies are fashioned. 

Figure 2.2. Merrily We Roll Along. 

Experiment 2: Tempo and Melody 

Another determining factor in how a melody will sound involves the 
speed/tempo at which the melody is played. Set the controls of the 
Odyssey as you did in the previous experiment and play the melody 
shown in Figure 2.3. Now play the melody as fast as you can at least 
two or three times, listening carefully as you play; then do just the 
opposite and play the music very slowly. Does the character of the 
melody change as the tempo changes? 

If you wish, expand this concept of tempo and its effect on melody 
by playing a lovely ballad — choose one of your favorites — and play 
it at an unreasonably fast tempo. Do just the reverse with a compo- 
sition normally performed rapidly — "Deck the Halls," for example — 
and play it very slowly. Regardless of how accurately the notes 
(pitches) are played, the change of tempo in instances such as these 
has a dramatic influence on how the melody is perceived. 


1 C3 


t ||, 

1 1- 




I a £ 



T Stir 

L £3C 

ui 2 

8 jt 

< O w 

5 £ 3 




P S i \ 





t « 


i 1 1 r 

=*1 5^ 

i| mi 



1 1 1 

0 0 ' 

J * J 

- 1 0 

Figure 2.3. Long, Long Ago. 

Experiment 3: Tonality and Melody 

To the two aspects of melody already discussed, rhythm and tempo, 
let's add a third: tonality. The intervalic relationship of the series of 
pitches that make up a melody determines the prevailing tonality of 
that melody. This concept is more easily demonstrated than verbalized. 
To try altering the tonality of a familiar melody, first set up the patch 
shown in Figure 2.4. 

The first musical example, shown in Figure 2.5, is the familiar 

Figure 2.5. Merrily We Roll Along. 

1 — "j — 



igure 2.( 

Minor-chord versi 

on of Figui 

•e 2.5. 





II I. 



h K J 




O j £t 

g |5 


-i 1 1- 

i i 1 - 


•HI mil 

Hi jcKi 





— ( l ! !! !>! sm^ 


child's tune, "Merrily We Roll Along." You've already played this 
melody in Experiment 1, at which time you discovered the effect that 
changing the rhythm can have upon melody. This time play the melody 
as modified in Figure 2.6. As you play, note the difference that the 
change in tonality makes. 

When you compare the two examples, you'll undoubtedly discover 
that the melody takes on a completely different character when the 
tonality is changed. If you have a knowledge of chords, you might try 
playing the first melody accompanied by major chords, the second by 
minor chords. This will augment the effect of the tonality change. 
Don't hesitate to experiment further by trying tonality changes in 
other familiar melodies. 

Experiment 4: Timbre and Melody 

The final experiment demonstrates how timbre interacts with mel- 
ody to create a variety of different melodic effects. You can exercise 
almost total control over the timbre of a melody when playing it upon 
the Odyssey; this control — through the selection of waveforms, the 
use of the VCF and HPF, and the settings of the AR and/or ADSR 
generators — permits you to simulate traditional instrumental timbres 
or to go far beyond their bounds. The point is that you should think 
about playing expressively and making full use of the controllable 
synthesizer functions when playing the Odyssey, just as a trumpet 
player can add expression to a melody by permitting his tone to 
grow brighter during sustained notes. In each case, the expressive 
interpretation of the melody, whether on a horn or on the Odyssey, 
results in a more interesting musical experience for both the player 
and the listener. 

To demonstrate the effects that a change in timbre (timbre being 
all those characteristics which distinguish any particular sound from 
another sound) set up the patch shown in Figure 2.7. Let's first try 
changing the timbre of this sound by modifying the attack and decay 
characteristics. Using the patch as shown, play the melody illustrated 
in Figure 2.8. Listen carefully as you play, so that you'll recall the 
basic effect of this melody as you play the next version. 

Now, reset the ADSR controls to different positions, playing the 
melody again with each new setting. You'll find that raising the attack 
slider to its uppermost position makes it virtually impossible to play 
a composition with notes coming in rapid succession — certainly a 
change in the melodic character. Try each of the envelopes in Figure 
2.9. Try experimenting with different settings of the VCF Resonance 
control and Portamento control, too. 


r^i It 
2< « 

l }\ \t 

f * 5 

tr uj > 

i| 1 



1 r- 



1 c 


S B * 


O § E 



2 1 


• HI mil 

* 5 <, 


8 . .i u 

If £±*z 

1 1 !i£ 

5 z u 
> o « 
o o 



- ::• i 



^ J J J. 

Figure 2.8. Twinkle, Twinkle, Little Star. 




i r 


- - 1 

I ] 


i : 

■- 1 

t 4 

J 1 

r -. 

: V 

I ADSR 1 | ADSR . 



Figure 2.9. Variations on setting in Figure 2.7. 

A similar experiment can be performed by recording a friend who 
plays the trumpet. Turn on the recorder and have him play a single 
sustained note, approximately five seconds in length. Then edit and 
splice the tape so that the attack and decay of the tone are elimi- 


nated — take only the central portion of the piece of tape, approxi- 
mately two or three seconds. Play back the result and listen to the 
difference. If possible, try this experiment with other instruments — 
a clarinet, a violin, even a tuba. In some instances you'll find that the 
net effect is startling — the instrument is hardly recognizable. 

At this point you may also wish to review the information provided 
in Part II on the envelope generators. This material will indirectly 
suggest many ways in which the character of the melody in Figure 2.8 
may be altered. By all means try the ADSR control of the pulse width 
and ADSR-controlled frequency modulation. The dynamic-waveform 
concept and its musical usefulness in terms of melody will become 
readily apparent. 

In addition to the above experiments, you should also try using 
several different pulse widths, from 10% to 50% (a square wave), as 
well as the sawtooth wave. Also try adding portamento and vibrato 
as you experiment (see Part II for patch settings, if necessary). The 
interaction of these effects and the effect upon the melody will come 
into clear focus as you experiment. 

While this discussion has systematically examined a variety of ways 
to alter melody, the total musical effect of the melody often depends 
upon all of the above considerations. The performer has the final 
say as to the particular components to be highlighted, subject to the 
limitations of the instrument he plays. 

Through this series of experiments, you have almost certainly come 
to realize that arrangers are very much aware of specific instruments 
and how they can be used most effectively to enrich melodic lines. 
This is orchestration. For instance, you'll rarely hear a tuba playing a 
love theme with a string orchestra; nor, on the other hand, would a 
violin play the foundational, or bass, part. Also, it's safe to assume 
that the "Minute Waltz" is not a favorite piece for trombone, while 
a pianist cannot produce a beautiful portamento effect on the piano. 
It follows that the timbral assets and limitations of an instrument are 
important reasons why an arranger chooses one instrument over 
another to play a particular part. 

As a conclusion to this discussion of melody, try "orchestrating" 
a number of familiar melodies, using several of the instrumental 
patches that appeared in Section 1 of Part III. Decide which instru- 
ments) you find most appropriate for a lullaby, which for a pop tune, 
and which for a musical television commercial most students have 
heard. While you may not agree on one single instrument that sounds 
best, you'll almost undoubtedly agree on two or three that are totally 


Section 3: Harmony 

In contrast to melody, which we established as an ordered series 
of pitches making up the basic theme or "tune" of a song, harmony 
is present as musical tones which are sounded simultaneously. There- 
fore, whenever you play two or more notes of different pitches, har- 
mony is automatically implied. As notes are sounded simultaneously to 
form various intervals (two notes) and chords (three or more notes), 
the resulting harmony further defines the tonality of a composition. 
More important, however, than any technical definition of harmony 
is its function: that is, to provide support for the melody, ranging 
from one simple counter-melodic line up to and including complex 
series of chords. 

Because almost all music is dependent upon intervals and the re- 
lationships between intervals, this section of the text is devoted to a 
systematic study of intervals, how they contribute to harmony, what 
is meant by being "in tune" or "out of tune," and how you may im- 
prove your own ear for harmony and frequency relationships. 

Experiment 1: Ear Training 

The ability to distinguish minute differences in pitch is almost 
indispensable to any musician, regardless of the kind of musical activ- 
ity in which he may be engaged. First, playing in tune is imperative 
for those persons who play tuneable instruments — strings, reeds, 
and brass instruments, for example. Musicians playing fixed-pitch 
instruments, such as the piano, also have need for a well-trained ear, 
however; accurately determining chord changes without music, re- 
hearsing singers, and generally filling the role of musical director, as 
the keyboard player often does, all require attentiveness to intervalic 
and harmonic relationships. 

The Odyssey is an excellent tool for providing ear-training experi- 
ences, as its infinitely variable, wide-range oscillators permit virtually 
any tuning. In addition, the VCF will be employed later in this section 
to "pick apart" the harmonics common to various tunings, dramati- 
cally demonstrating relationships between various intervals that could 
only be talked about theoretically without the use of a synthesizer. 

Begin Experiment 1 by setting up the patch shown on Figure 3.1. 
Having now established a basic frequency setting on VCO-1, press 
the key on the Odyssey keyboard indicated by a circle in Figure 3.2. 
While this pitch continues to sound, raise the VCO-2 attenuator into 
the Audio Mixer and tune VCO-2 to the same pitch as VCO-1. To do 
so, you should first move the coarse-tuning slider to approximately 
the same position as the corresponding slider on VCO-1; then, continue 


Figure 3.2. Tuning unisons. 

tuning by adjusting the fine-tune slider on VCO-2. When the two 
tones are nearly in tune, you will hear a pronounced wavering sound, 
or "beat" — the tones will sound as if they are growing louder and 
softer. The speed of the beat depends upon how far out of tune the 
oscillators are. The closer they are to being in tune, the slower the 
beat will become. Continue to tune until the beats finally slow down 
and then stop; the pitches/frequencies of VCO-1 and VCO-2 are now 
identical, or in tune. In this instance, they are tuned in unison, or to 
the same pitch. 

The reason that you hear beats when the oscillators are slightly out 
of tune is because two waveforms that are at slightly different fre- 
quencies produce an alternate cancelling effect, thus causing a cyclic 
change in amplitude, resulting in beats. This effect is a useful one, 
since it not only aids us in tuning unisons accurately, but also will 
permit the accurate tuning of other intervals containing common 
harmonics. While the fundamental frequencies of two tones may be 
different, if they have harmonics in common, these harmonics will 
beat as the interval being tuned is brought nearer and nearer to being 
in tune. 

Now continue this exercise in ear training by tuning the oscillators 
to an interval of a major third. This commonly used musical interval 
is represented on the keyboard in Figure 3.3 by the circle and the tri- 
angle; the key upon which the triangle is placed is a major third 
above the basic unison pitch already tuned. 

To tune this interval, listen first to the pitch produced when the 
key with the circle on it has been pressed down. Then press the tri- 
angle-coded key. Alternate between the two until, while listening to 
first pitch, you can easily hum the second pitch (that of the key with 
the triangle on it). When you can do this, tune VCO-2 to the pitch 
you are humming. Note: you should not attempt to do this with the 
coarse-tuning slider, as it will take you beyond the interval you want 


ill So_ 

i IJCOii 


*1 k : : : * 

5 2 S 


O H K 


SS I 1 1 1 

— ' — r- 



O <r 

• cm 

*Hf mil 

i £ 





1 x 

" z o 

• fi S 

3»: 2 OZlf 

S 2 
X. s 


Figure 3.3. A major third. 

to tune very quickly. Instead, use the fine-tune slider of VCO-2, 
raising it slowly until you hear the pitch of the major third come 
into tune. 

Use the same procedure to tune the major fifth, using the keys indi- 
cated by the circle and the square in Figure 3.4. Listen for the beat 
of the harmonics the two frequencies have in common. You should 
also practice tuning two additional intervals: a fourth, indicated in 
Figure 3.5 by the circle and the star, and an octave, indicated by the 
first circle and last two circles in the same illustration. 

Figure 3.4. A major fifth. 

Figure 3.5. A fourth and an octave. 


Experiment 2: Using the VCF to Listen to Specific Harmonics 

Without the use of an electronic music synthesizer, the study of 
harmonics is one which has had to be taken pretty much on faith 
alone. Certainly, theory existed which seemed to demonstrate the 
presence of harmonics and their effect upon the timbre of the sounds 
you hear. Helmholtz first supplied the theoretical basis for such an 
understanding. To illustrate, let's use a vibrating string, as shown in 
Figure 3.6. It has been proven that a vibrating body, such as the string 
shown in this figure, does much more than simply vibrate back and 


Figure 3.6. Vibrating string. 

forth as a whole. Indeed, simultaneous but distinctly separate vibra- 
tions occur, so that the string is also vibrating in sections of one-half, 
one-third, one-fourth, and so on. Each of the vibrations, and the fre- 
quencies the smaller sections provide, becomes weaker as the divisions 
of the string become smaller. The pitch that is most prominent — that 
pitch resulting from the first and largest motion of vibration — has 
the greatest amplitude and is called the fundamental. This, of course, 
provides the basic pitch of the note that you hear. All of the fre- 
quencies produced by smaller divisions of the string are called over- 
tones, or harmonics. 


While this information has been known since the latter half of the 
19th century, it has never been something that could be easily dem- 
onstrated. Through the use of the Odyssey's low-pass and high-pass 
filters, however, it is now possible to single out virtually any harmonic 
component of a waveform, dramatically demonstrating its presence. 
It is also possible, when tuning intervals other than unison, to sweep, 
or scan, the harmonics of both waves — picking out common harmon- 
ics having a potential beat (if the interval is slightly out of tune), thus 
demonstrating why you perceive a beat when tuning an interval such as 
a fifth. Since it is not possible for the fundamentals to beat, since they 
are two different frequencies, any beat perceived when tuning intervals 
other than unison must be. because of harmonics held in common 
between the two notes. The following exercises will graphically dem- 
onstrate this fact. 

First, set up the patch shown in Figure 3.7. Note that the resonance 
on the VCF is very high, in order to elicit a peaked response or band- 
pass effect around the narrow band of frequencies determined by 
the setting of the initial filter frequency slider. Now, by raising the 
initial filter frequency slider (marked VCF Freq), you will be able to 
scan, or pick out, the harmonic components of the waveform being 
sent to the filter by VCO-1. The pitches of the harmonics you will 
hear, from the fundamental up through the ninth harmonic, are shown 
on the staff in Figure 3.8. 




Figure 3.8. First nine harmonics of a pulse wave. 




l irpg l 

I *JL 


x5ES 1 1- 


1 «s 

2 < ™ 
i a i 

£ of! 



1 1 r- 

j i 1— 

■ H 

o c 


*H! mis 


I ||C 


J t 

\v\ i : ; » h ots 

s*> 3 o 



> s r- 

O E « 

£ 5 <tZ 


i t 


O z o 

> o » 

1 — I 1 I I 






Having now heard the fundamental and all of the harmonics of the 
basic waveform, tune VCO-2 to unison. When you have achieved a good 
unison tuning, again gradually raise the VCF Freq slider, picking out 
the fundamental, then the second harmonic, then the third harmonic, 
and so on. If you hear a beat at any point, this means that a perfect 
unison has not been achieved and you should sharpen your tuning. 

After scanning the harmonics of the unison waveforms with the 
oscillators in tune, then slightly detune VCO-2 by moving the fine- 
tuning slider up just slightly. With the VCF Freq slider open most of 
the way, you should be able to detune easily to get a perceptible beat 
of one or two cycles per second. With the oscillators detuned in this 
manner, scan the harmonics again. You'll note that with unison tuning, 
the fundamental and every harmonic will beat. Logically, you will 
understand that this is as it should be, since two fundamentals of 
exactly the same frequency will have the same harmonics. Figure 3.9 
illustrates this. 

n i n — i 

4« — 










Figure 3.9. Harmonics in unison tuning. 

Now tune VCO-2 a fifth above VCO-1, as you did in Experiment 1 
of this section. When you have achieved a good tuning, again detune 
slightly so that you get a perceptible beat. As discussed earlier, the 
beat you are now hearing is the result of the harmonics that the two 
frequencies have in common (Figure 3.10). 


1 q Rpat.s . — q 

&^ — — ^ 

— 3^1 


ats £J 



— & — 

Figure 3.10. Harmonics when VCO-1 and VCO-2 are a fifth apart. 

If you now scan the harmonics, again using the low-pass VCF, you 
will find that every harmonic does not beat, as it did when the oscil- 
lators were tuned to unison. This is because with a fundamental fre- 
quency of VCO-1 being different from the fundamental frequency of 
VCO-2, all of the resulting harmonics are not common. As you can 
demonstrate by scanning with the VCF, every third harmonic of VCO-1 
will beat with every second harmonic of VCO-2. Figure 3.11 indicates 
why this is so; if a note appears on the same line for both VCO-1 and 
VCO-2, a potential beat exists between those harmonics if the interval 
is not perfectly tuned. On the other hand, on those lines and spaces of 
the staff where VCO-1 and VCO-2 do not both have a note, indicating a 
harmonic, there is no beat (potential or otherwise) and you should hear 
a steady, even harmonic pitch. 

The chart in Figure 3.11 illustrates the same concept, but uses 
arbitrary frequencies of 100 Hz (VCO-1) and 150 Hz (VCO-2) for 
the purposes of numerical simplicity; the relationship between these 
two round numbers, however, is the same as that of any particular 
pitch and that pitch a fifth above it. Therefore, the harmonic relation- 
ships expressed in the chart will hold true whether you tune the first 
oscillator to middle C, A-440, or any other frequency. 

VCO-1 (100 Hz) 
Potential Harmonics 

VCO-2 (150 Hz) 
Potential Harmonics 










Figure 3.11. Frequencies of harmonics of two pitches a fifth apart. 


Experiment 3: Using the High-Pass Filter to Eliminate Certain 

A kind of reverse experiment can be performed by using the 
Odyssey's HPF (High-Pass Filter). In this experiment, you will raise 
the HPF Cutoff Freq slider in order to eliminate the harmonics from 
"the bottom up." Set up the patch shown in Figure 3.12. You'll hear 
a continuous tone, the sound of the narrow pulse wave of VCO-1. 
Now, gradually raise the HPF slider, listening carefully as each of the 
harmonics drops out. The first to go, of course, should be the funda- 
mental — the basic pitch itself. 

The musical example shown in Figure 3.13 illustrates what is hap- 
pening in terms of the fundamental and all harmonics shown on a 
staff. Note that an interesting effect occurs after you have filtered out 
the fundamental frequency; despite the fact that you have filtered 
the frequency which provides/provided the basic pitch reference, it is 

High-pass Filter 
Cutoff Point 







F. 2345678 9. 

Figure 3.13. What happens when harmonics are filtered. 

still implied — you are still hearing the original pitch determined by 
the setting of the oscillator and the last note pressed on the keyboard. 
You can go even further and remove many of the lower harmonics 
without altering this. This effect is called residual pitch. If this tone 
is interrupted, however, and then resumed — as it might be if you 
closed the VCA, and then returned to the Odyssey five minutes later 
and reopened it — the sensation of pitch will be completely altered. 
Instead of hearing the residual pitch based upon the original funda- 
mental frequency, you will hear the formant pitch, which will lie 
approximately in the region of the strongest remaining harmonics. 


r it 

II I *" 



O 5 E 

£ £ 3 

5 * 'I 





j — i — i — k. 

41 j I3ppsi*l 










Experiment 4: Counter-Melody 

You'll recall that in the beginning of this section we said that 
harmony is implied any time that two or more notes are sounded. 
Counter-melody is an excellent demonstration of this fact, providing 
the simplest of exercises in two-note harmony. Many of the "rounds" 
that you sang as a child depend upon this kind of effect. 

The round shown in Figure 3.14 will provide an adequate demon- 
stration. This music may be performed in two ways: if you are using 
the Odyssey with a tape recorder, record the first part (Part A); then, 
as the recorded portion is being played back, play Part B along with 
the tape. If you are not using tape equipment, you can simply have 
one student play Part A in the lowest octave of the Odyssey's key- 
board while a second plays Part B in the highest octave. 



9 4 


Figure 3.14. Row, Row, Row Your Boat. 


Experiment 5: Two-Note Chords with a Recorded Melody 

The final experiment in this section will provide an experience in 
harmonizing a simple melody with two-note chords played on the 
Odyssey. You will find that, despite the fact the melody is the same 
in both instances, the interrelation of melody and harmony is such 
that if the tonality of the harmony is changed, the net effect of the 
music upon the listener is greatly altered. 

If you are using tape equipment, you should first record the melody 
(upper line) shown in the two examples; this is exactly the same in 
Version A and in Version B. Then, as the recorded portion is played 
back, play the harmony part (lower two-note part) indicated in Ver- 
sion A. After you have had a chance to hear how this harmony sounds, 
play the part indicated in Version B. 

* 1 


0 : 


Figure 3.15. Aura Lee. 





— o 


fp: p — 




, hg 

Figure 3.16. Minor-key version of Figure 3.15. 

When you compare the two examples, you'll probably discover that 
the melody takes on a completely different character when the tonality 
of the harmony is changed. The first example is harmonized in a 
major key; the second is in a minor key. The minor chords providing 
the harmonic background in Version B typically create an effect 
best described as "dark" in nature, and tending to convey feelings of 
sadness. Major keys, on the other hand, tend to provide a "lightness" 
characteristic of Version A. While these two general descriptions are 
arbitrary, and probably stem from our experience — associating minor 
keys with funerals and blues — they are set forth here simply to pro- 
vide an easy "handle" by which to describe the effect that the har- 
mony has upon the impact of the musical whole. Don't hesitate to 
try reharmonizing other melodies that members of the class know well. 


Section 4: Transposition 

There are many ways of classifying musical instruments. One such 
classification, mentioned in Part II, divides them into two general 
categories: (1) concert-pitch instruments, and (2) transposing instru- 
ments. The concert-pitch instruments provide the same sound notated 
on the music. In other words, a particular pitch/note is produced by a 
key which has the same name. On the other hand, a transposing 
instrument is one in which the player reads one note and the instru- 
ment sounds another. Therefore, musical notation for a transposing 
instrument has to be written in another key; its pitch will then match 
those instruments tuned to A-440, or concert pitch. As you can well 
understand, this basic difference in the pitch produced by concert- 
pitch and transposing instruments can and has caused certain problems. 
For example, if a clarinetist (B-flat instrument) wanted to play a flute 
part (C instrument), one of two things would be required: (1) trans- 
posing the music by rewriting the part, or (2) mentally reading the 
notes in another key. 

With a few exceptions, most brass and woodwind instruments are 
not tuned to concert pitch, while keyboard instruments — piano, 
organ, accordion, and celeste — are tuned to concert pitch. This pitch 
difference in the two kinds of instruments makes it impossible for 
them to be played together unless one of the above-mentioned con- 
siderations is met. In the normal course of events, the first considera- 
tion — written transposition — is automatically met in the scoring of 
the music by the arranger, thus eliminating this step by the player. 
The transpositions in the score, however, would only cover those 
normally needed and still leave the clarinetist in our example relying 
on personal devices to play the flute part. Keyboard players desiring 
to play parts written for transposing instruments could also expect to 
deal with the same problems. As can easily be seen, the interchanging 
of one instrument for another has important limitations other than 
just the differences in sound/tone color which two instruments 

By now you may have wondered why all instruments are not tuned 
in the same key — a question that many an individual struggling to 
transpose a part has thought about. The reason can be explained by 
very briefly examining the history of the early development of musical 

Originally, the musical staff notation used by an individual to play 
a particular instrument was determined by the range of the instru- 
ment — in this case the lowest note. If the lowest clear tone that an 
instrument could produce was a B flat — using today's pitch stan- 


dards — then the instrument was said to be pitched in B flat and be- 
came known as a B-flat instrument. These early instruments did not 
have the benefit of valves and slides and therefore could not duplicate 
the range of their modern-day equivalents. Although valves and slides 
have extended the capability of instruments so that they can play 
effectively in almost any key, transposed notation has survived and 
become a tradition. 

The result of these differences in instrumental pitch is a musical 
score which by necessity contains various key signatures to accomo- 
date these pitch differences. Examine Figure 4.1 and locate two instru- 
ments with unlike key signatures. 

Figure 4.1. Example of an orchestral score. 



Find a simple melody and transpose the music for the trumpet. 
Remember, the notes must be written one whole step higher. 


Alto Saxophone 

The alto saxophone is another transposing instrument. Unlike the 
trumpet, the alto saxophone is an E-flat horn producing a note 4V6 
steps lower than the written notation (Figure 4.3). Just a moment's 
glance at Figure 4.1 provides an immediate indication of one area in 
which a conductor must be extremely proficient — transposition. 
It's readily apparent that the study of musical scores remains a dif- 
ficult task for many musicians, the beginners and the experienced 
alike. To rehearse and conduct an orchestra competently demands a 
thorough knowledge of transposition on the director's part, and the 
concepts of transposition continue to challenge most musicians. 

Transposing and Concert-Pitch Instruments 

Now that you have acquired a general idea of transposition and its 
over-all importance, let's examine some specific instruments in terms 
of the pitch they produce. 


The trumpet depends on transposition to produce notes in the 
concert key. It is pitched in B flat, which means that the trumpet 
part must be written one whole step higher to facilitate the correct 
adjustment in pitch (Figure 4.2). Transpose a simple melody for the 
alto saxophone. You may find this transposition more difficult due 
to the wider interval. Check the notes carefully. 


The trombone is an instrument that does not need to be trans- 
posed — it produces the notated pitch. Figure 4.4 shows trombone 

Figure 4.4. Trombone part example. 


Another example of a concert-pitch instrument is the flute. Figure 
4.5 illustrates flute notation. 

At this juncture you've learned that pitch differences between in- 
struments, which made the concept of transposition necessary, was 
brought about by the limitations of early instruments and continued 
because of tradition. Since it doesn't appear that our notation system 


Figure 4.5. Flute part example. 

is going to change, pitch alterations by transposing will undoubtedly 
remain the usual practice. Therefore, all the skills needed to make this 
system operational will also be required. There is an instrument with 
the capability of instantly mastering any transpositional situation, 
however. Imagine playing an instrument which functions in a manner 
so sophisticated that it immediately allows you to transpose to the 
key of your choice without fulfilling any of the usual steps. This in- 
strument is, of course, the Odyssey (Figure 4.6). 

Figure 4.6. The ARP Odyssey. 


Now, let's take an example — the clarinet — and perform the neces- 
sary steps to allow a band or orchestra member to assume the respon- 
sibility for playing a clarinet part on the Odyssey. 

The first step is to set the controls to create the sound of the desired 
instrument. Figure 4.7 shows the patch for the clarinet sound. Second, 
determine the interval difference between the note read and the note 
produced. Figure 4.8 follows through with the clarinet example. Use 




Figure 4.8. Clarinet transposition. 

this same format for assessing the pitch differential between other 
instruments. Initially, if the Odyssey is tuned to concert pitch (A-440) 
then A played on the Odyssey keyboard will produce the same pitch — 
A. Figure 4.9 illustrates. 

The second step in Figure 4.9 clearly shows that the concert-pitch 
tuning will not be acceptable. Therefore, move the fine-tuning slider 
of VCO-2 until the pitch becomes one whole step lower. The Odyssey 


Figure 4.9. Concert-pitch tuning. 

is now transposed to the Key of B flat — playing C on the keyboard, 
produces a pitch one step lower or B flat. Figure 4.10 shows this rela- 
tionship. Double check your tuning by trying the following example: 



Play G on the keyboard. What note should this key produce? Figure 
4.11 answers the question. 


p — 






Figure 4.10, Figure 4.11. Transposed tuning. 

In addition to adjusting the keyboard pitch to facilitate playing 
various instrumental parts, another important consideration enters 
the picture — the range in which an instrument plays. The Odyssey 
control employed to satisfy the particular range requirements of each 
instrument — the Transpose switch — is shown in Figure 4.12. To 
demonstrate the range capability of the Odyssey, first press the circled 
key shown in Figure 4.13. Now play the key indicated with a triangle 
two octaves lower. Remember the sound. 

Play the first key (circled letter) again. This time move the trans- 
pose switch to the "down" position, and play the same key. Did you 
hear the identical pitch produced by the key labeled with the triangle? 
Perform a similar experiment, this time starting at the lower end of the 
keyboard, and produce a note two octaves higher. Exploring the total 
range possibilities of the Odyssey will yield seven complete octaves — 
ample flexibility for the creation of almost any instrumental sound. 

Figure 4.13. Two-octave transposition. 


p p|q:li 

?~ {iJlZIZIgg 

j$ I CO|<( 


1 //f^i 





Tape Music Techniques 

The rapid proliferation of electronic music synthesizers has only 
served to further accelerate the already impressive growth of contem- 
porary recording arts. Electronic music studios have popped up as 
natural additions to long-established recording studios; new demands, 
in turn brought about by the increasing number of composers work- 
ing in this medium, have inspired exciting new recording techniques 
and equipment. Equally important, these composers have encouraged 
all of us concerned with, and involved in, the making of music to look 
at tape recording in a new light. The situation where the performer 
goes into a studio to record the work written and orchestrated by the 
composer has changed; instead, it is likely to be the composer who 
goes into the studio and performs his own work, meticulously building 
layer upon layer of sound until, at the end of the session, the com- 
poser-performer is able to walk out of the studio with his compo- 
sition-performance on a reel of tape in his hand. Composition and 
performance have been fused into one act, through the use of elec- 
tronic music synthesizers and sophisticated recording techniques. 

The reasons for the emergence of the composer-performer are 
probably as numerous as the motivations which cause men to write 
music. Certainly, the old dichotomy of the composer's intentions 
versus the performer's interpretation is neatly resolved by this method. 
More significant, however, is the fact that the composer is no longer 
concerned with just melody, harmony, and rhythm; a fourth element 
has been added — timbre. Electronic music synthesizers have opened 
the door to infinitely broad timbral resources, resources which con- 
temporary composers are seeking to exploit to the fullest. No longer 
is the composer bound by the timbres which could be produced by 
conventional musical instruments. With that freedom, however, has 
come the concomitant problem — how does one notate the subtle 
changes in timbre, much less the all-encompassing audio fabric, that 
is woven from single strands of synthesized sound? Music notation, 
which has changed in the past to accommodate new forms of musical 
expression, undoubtedly will change in the future to include the ad- 
vances we view as radical today. Nevertheless, until that time comes, 
the composer must almost inevitably become the performer, and not 
surprisingly, many are going to be reluctant to relinquish this new- 
found role when the moment does arrive. Therefore, it seems logical 
to expect that fusion of the roles of composer and performer is likely 
to herald a new era of musical endeavor that will not be totally revers- 
ed, despite the ultimate development of adequate notation systems. 

It is for this reason that throughout this text, mention is made of 


the use of tape equipment. Not only is the joint role of composer- 
performer a pleasurable one, it is an almost essential one — particularly 
if you are to preserve the musical expressions you create when using 
the Odyssey. Therefore, it is suggested that whenever possible, the 
electronic music studio that is developed around the synthesizer 
should include at least one good-quality tape recorder. If your school 
already has such a machine, this section of the text will provide sug- 
gestions as to how you can make the most of it; if you do not present- 
ly have access to a recorder, use the guidelines in this section as the 
basis for making a decision when the opportunity to acquire such a 
machine arises. 

Section 5: Setting up an Electronic Music Studio 

An electronic music studio need not require thousands of dollars 
worth of expensive electronic equipment. Indeed, the ARP Odyssey 
will provide the basis for a sound, well-equipped studio, and all other 
equipment can be added as the budget permits. Most studios are begun 
in whatever room is available — a portion of the music room, a practice 
room, or an unused office. The following basic list of equipment may 
be helpful. 

Electrical Outlets. All that is necessary here is to make sure that the 
number of outlets is sufficient to handle the synthesizer and whatever 
auxilary equipment is to be used — recorders, amplifiers, and so forth. 
Outlets should be of the three-pronged type, to insure that each elec- 
trical unit is adequately grounded. 

Tables. The studio should accommodate a sufficient number of 
tables for the synthesizer(s), recorder(s), and other equipment that 
will be used. Naturally, all tables should be as sturdy as possible to 
prevent damage to any equipment. Shelving for storing tapes, etc., 
may also be installed after the tables have been placed. 

Bulletin Board. A bulletin board will be useful in maintaining a 
schedule for use of the studio and/or the equipment contained therein. 
Also listed on the board should be, any checklists and procedures for 
equipment operation. One word of caution, however: do not use 
chalkboards in the room; the chalk dust can cause expensive problems 
if allowed to accumulate on recording equipment. Use a board with 
pushpins for attaching notes and checklists, or secure one of the new- 
types of message boards that utilize a wax crayon on a plastic finish. 
These will wipe clean with no dust. 

Tape Recorders. While not essential, a recorder will add greatly to 
the enjoyment and utilization of the Odyssey. More detail will be 
provided as to the most useful type of recorder later in this section. 


An AmphterApeaker System. The Odyssey contains no speakers 

ava hlJV " h " y C ° mPatibIe With alm ° St ^ equipment that is 
available. Frequently, a portable public address system used for as- 

employed. Many schools also own either electronic pianos or an elec- 
tronic organ into which the Odyssey can be connected. You may also 
be able to connect the Odyssey to the external input of a recorder or a 
record player (assuming that the recorder or phonograph contains 
speakers of its own). Figure 5.1 shows a number of simple connec- 
tions; depending upon what type of equipment is being paired to the 
Odyssey a notation has been made on the back panel of the Odyssey 
as to which output to use - the high-level or the low-level output 

^^th m ^° m . SPedal taPG teChniqUeS may ™ c ™^ 
altering these basic connections. 


Use low-level output. 
Similar connection for 
portable p. a. might 
require high-level 



Use high-level output. 
Use Y- jack if two 
speakers are available. 








\ . 

Figure 5.1. Five possible connections for use with the Odyssey. 




Use high-level output. 
Use Y-jack to provide 
signal to both channels. 



Stereo Amplifier 

Turn Table 





Use low-level output 
to amplifier. Use high- 
level direct to tape 
recorder, with no 






Use high-level output 
with Y-jack to provide 
sound for both channels. 

Figure 5.1, continued. Five possible connections. 


In the event that you will be using a tape recorder, the following 
supplies should also be available. 

Tape. Fresh, unused tape should always be available for serious 
studio work; practice and experimental work can be performed on 
tape that has been used and erased. Avoid extra-long reels of tape; 
the longer the tape that is wound upon a reel, the thinner it must be 
to get the extra playing time on that same reel. Use IV2 mil tape; it 
will be easier to handle and less prone to stretching and accidental 

Splicer. The use of a tape splicer is both easier, and safer, than the 
use of a razor blade. Do not use magnetized scissors to cut tape; not 
only will they do a sloppy job but the magnetized blades will create a 
"pop" on the tape. 

Cleaning: Recorders need periodic care in order to operate at their 
best. However, the cleaning and demagnetizing of recorder heads 
should be performed only by those persons with appropriate training 
and experience. 

The one item not yet mentioned, of course, is the synthesizer 
itself. The ARP Odyssey is just one of a full line of electronic music 
synthesizers made available by ARP. All ARP equipment is compatible, 
and the studio that begins with an Odyssey can later add its big 
brother, the ARP 2600, or, bigger yet, the ARP 2500 modular system. 
Product information on each of these instruments is available directly 
from ARP Instruments, Inc., Newton, Massachusetts. 

Using the Odyssey with a Recorder 

As already mentioned, use of a tape recorder will greatly enhance 
the value and enjoyment of your experience with electronic music. 
Even the simplest monaural recorder offers at least the opportunity 
to record one part and then play it back while performing a second 
part on the Odyssey. The following information details how you can 
get the maximum use from whatever equipment is available, from a 
monaural machine up to and including a four-channel recorder. 

Monaural Machines. 

With a mono recorder, the possibilities are somewhat limited; for 
this reason, if you are about to acquire a new machine, it is suggested 
that something with greater flexibility be considered. Nevertheless, a 
monaural recorder can be used to create a basic library of taped sounds 
that will provide an audible record of experiments and patches, as 
well as the beginning of a sounds-and-effects library that may be used 
to contribute to the soundtrack accompanying school drama events. 


With a mono machine, you have two options: first, you can record 
and simply listen; this would include recording and storing those tapes 
of the best patches, effects, and experiments. Second, you can record 
one part, a melody, for example, and then play the recording while 
adding a second part, perhaps a counter-melody. Use of a Y-jack on 
the input of the machine will permit you to record both the Odyssey 
and the signal from a microphone, phonograph, or another recorder. 
In the end, however, you will find yourself constantly improvising 
connections in an effort to make the machine perform functions for 
which it was not designed. 

Stereo Tape Recorders 

The availability of a stereo recorder is a big plus for an electronic 
music studio. While not the ultimate machine, it is easily within the 
reach of most budgets, and a variety of good recorders are readily 
available. Most stereo machines have a "sound-on-sound" provision 
that will allow you to record one channel, then play back that chan- 
nel while at the same time recording a second part — in effect, adding 
a second sound upon the first sound. This provision will open the door 
to multitracking, in a basic sort of way, permitting actual composition 
using the Odyssey and the recorder. 

Most recorders of this type will also permit you to go beyond just 
two tracks by using a "bounce" technique. As the name implies, you 
record on one channel and then bounce the signal from that channel, 
along with a new part, onto the other channel. You can then bounce 
the net result, along with yet another part performed on the Odyssey, 
back onto the original channel. Figure 5.2 illustrates. 

Part Being Performed 

Recorded on Channel 

No. 1 - Bass Part 


No. 2 - Rhythmic Effect, plus 
recorded bass from 
Channel A 


No. 3 — Chord Accompaniment, 
plus recorded bass and 
rhythm from Channel B 


Figure 5.2. Diagram of "bounce" technique. 



Left channel 


1). Record "Bass" part on 
Channel A (left). 


Left channel 

Right input 





2). Record 2nd part on Channel B 
(right) along with "Bass" part 
being played back from Channel A. 

Right output 
Left input 




3). Record 3rd part on 

left channel, mixed with 
output of right channel. 

Figure 5.2, continued. Connections for "bouncing" between tracks. 


The shortcomings of this procedure are that each successive bounce 
will result in some loss of signal quality, accompanied by an increase 
of tape noise. As a result, the number of bounces will be limited by the 
basic quality of the equipment you are using and the cleanliness and 
operating condition of its tape heads. Try this technique to see how 
many "generations" away from the original track you can go without 
a significant loss of fidelity. Each time that first bit of recorded ma- 
terial is transferred to another channel, it is said to be one generation 
further away from first generation. Therefore, when it is bounced 
for the first time, onto channel B, the bass part in Figure 5.2 is al- 
ready second generation. When it goes back to channel A, with the 
added material from recorded parts two and three, it is into its third 
generation. If you were then to add a melody part, through one more 
bounce back to channel B, your original track would be into its 
fourth generation. At that point, you will probably notice an obvious 
loss of quality. For this reason, you should always record the less criti- 
cal material first, with the most delicate or exacting material to be 
saved for last. If you find that your stereo recorder will permit you 
to record four separate lines, as outlined here, you will indeed be well 
on your way toward the exploration of the fun of tape composition. 

The only equipment other than the recorder and the Odyssey that 
should be required for such an exploration is a Y-jack or patch cord. 
This will permit you to combine the signal from the Odyssey with 
the output signal from the channel of the recorder that is being bounc- 
ed to the second channel. Thus, when performing part number 2 in 
Figure 5.2, the Y-jack should be taking the rhythmic effect from the 
Odyssey, along with the output from channel A, and feeding it into 
channel B. When you move on to part number 3, you'll be taking the 
signal from the Odyssey again, but this time you will be combining 
it with the output from channel B and feeding the Y-combined signals 
into channel A. Best results will require some practice, particularly 
in terms of the balance of the volume levels of the various parts, but 
the results will be more than worthwhile. 

A Four-Channel Recorder 

Short of acquiring prohibitively expensive, professional-quality 
studio equipment, a good-quality four-channel recorder is the ultimate 
machine for a basic electronic music studio. As the name implies, 
this recorder permits you to record four separate channels, all first 
generation, resulting in high-quality, multiple-track tape composition. 
Moreover, the more advanced machines, such as the TEAC 3340, en- 
able you to go beyond four channels to at least seven tracks, none of 


which will be more than second generation. The key to the flexibility 
of such machines is a "record head" that can also be used for monitor- 
ing. Before going any further, let's define what a "head" is and look 
at the normal configuration of tape heads. 

The tape heads are simply the parts of the machine that make con- 
tact with the recording tape, encoding (recording) and decoding 
(playback) the magnetic signal. Normally, there is also an "erase" head 
that precedes both the record and playback heads, in order to erase 
unwanted signals before the record head applies new magnetic signals 
to the tape. Looking head-on at a typical machine, the uncovered 
heads might be positioned something like the diagram in Figure 5.3. 





Figure 5.3. Standard tape recorder. 
On the less sophisticated recorders, the record head is not capable 
of monitoring tracks already recorded while at the same time record- 
ing those tracks still open. Instead, the recorded tracks are monitored 
by the playback head, resulting in a time lag between the signal being 
recorded on the second track(s) and the signal being monitored on 
the first track(s). Remember, the tape is moving across the heads at a 
particular speed measured in inches per second; the slower the tape 
moves, the more time that distance between the record and playback 
heads represents. At 3% ips, which is a common speed on stereo ma- 
chines, the time lag between heads can be as much as a quarter of a 
second. Imagine being a quarter of a second out of meter when playing 
with a group; if you can mentally hear the rhythmic "fight" that would 
result, you can begin to imagine the effect that would result from 
trying to record a new track in perfect synchronization with a track 
you're hearing a quarter of a second later than you should. 


For this reason, most modern quarter-track machines have the 
synchronization feature mentioned earlier: a record head that can 
also be used for monitoring, thus eliminating the time lag between 
heads. All legitimate four-channel machines also have all four channels 
running in the same direction. Most stereo machines actually record 
four separate channels, but can only record or play back two at a time. 
Such machines are called "quarter-track stereo" recorders. When the 
reels are switched and the tape is reversed on such machines, stereo 
recording continues in the other direction. This is potentially useful 
for getting more stereo material on a single reel of tape, but is not 
particularly advantageous in an electronic music studio (Figure 5.4). 

Must turn tape around. 







All four tracks, same direction 



Figure 5.4. Comparison of two types of four-track recorders. 

A four-channel, synchronized machine, then, permits the recording 
of four individual tracks with ease. One simply records track 1, then 
monitors 1 while recording 2, monitors 1 and 2 while recording 3, 
and finally, monitors 1, 2, and 3 while recording 4. See Figure 5.5. 

Musical Track 










Figure 5.5. Four-track recording. 


It is also possible, however, to get six or seven tracks, with no 
track being more than second generation (Figure 5.6). 

Musical Track 









Mix A+B+C 

D = 1+2+3 


A = 4 


B = 5 


C = 6 

Figure 5.6. 

Procedure for 
six-track recording. 

/ All 6 tracks 

To produce six tracks (three first generation and three second gen- 
eration), simply record tracks 1, 2, and 3, then mix and record them 
on D. This frees channels A, B, and C for musical tracks 4, 5, and 6. 

To achieve seven tracks, follow the procedure shown in Figure 5.7. 

Musical Track 









D = 1+2+3 






C = 4+5 


A = 6 


B = 7 

Figure 5.7. 

All 7 tracks 

Now your four-channel tape contains seven separate musical tracks: 

A = 6 

B= 7 

Mix = 1+2+3+4+5+6+7 

C = 4+5 

D = 1+2+3 



Record Track 1 



Record Track 2 














Record Track 3 








Mix A+B+C onto D 













Record Track 4 



Record Track 5 










Mix A+B onto C 



















Record Track 6 



Record Track 7 








Play all 7 tracks 


















► OUT 

Figure 5.7, continued. Connections for 7-track recording. 


The result shown in Figure 5.7 is seven separate tracks, five second 
generation and two first generation, resulting in extremely high musi- 
cal quality and ample possibilities for exploration of tape composition. 

Our constant concern for the quality of the recorded signal is re- 
flected in the development of eight-, sixteen-, and twenty-four-track 
professional machines which are now in use in studios around the 
world. These machines, such as the one pictured in the studio layout 
shown in Figure 5.8, offer as many as twenty-four individual tracks 
and provide virtually unlimited multitracking potential without going 
beyond first generation on any track. This is particularly useful in re- 
cording large orchestras, since it is not necessary to devote a separate 
channel to each instrument; rather, sections of instruments may be 
covered with one or two microphones, while other difficult-to-record 
instruments (such as a drum set) get one or more mikes of their own. 

Figure 5.8. 

If your electronic music studio includes a four-channel tape ma- 
chine, you'll find that its seven-track potential permits all but the most 
complex tape compositions to be recorded. In the event you need 
more than seven tracks, you may get up to eleven tracks on a four- 
channel recording by bouncing from track to track as shown in 
Figure 5.9. 


Musical Track 









D = 1+2+3 






C = 4+5 




A = 1+2+3+4+5+6 






D = 7+8 




B = 7+8+9 





Musical tracks 1, 2, 3, 4, 5, 7, 8 will be 3rd generation. 
Musical tracks 6, 9 will be 2nd generation. 

Musical tracks 10,11 will be 1st generation. 

Figure 5.9. Eleven-track recording. 

As in the case of the stereo machine, be certain to record less critical 
parts first. As you mix down a number of tracks onto a single channel, 
be careful to watch the volume levels of the individual tracks to be 
sure that you are getting the balance you want between the parts. Ex- 
perimentation and lots of practice will prove to be your best guide. 


Tape Editing 

Up to this point, this section has been primarily concerned with the 
kind of equipment that you may have available and how to maximize 
its potential in terms of the number of individual tracks that you may 
record. We have spent a considerable amount of time on this subject 
because it so directly affects the potential complexity of the tape 
compositions that you will be able to create. Nevertheless, there are a 
number of subjects that will prove to be of equal importance as you 
begin incorporating a recorder into the activities of the electronic 
music studio. One such subject is tape editing. 

Virtually every contemporary recording you hear is not the result 
of a single continuous "take," but rather is the best parts of several 
tries, or takes. The reason that finished recordings are assembled from 
the best pieces of several takes is largely because of the conven- 
ience to the performer and the lower cost in time and labor. While a 
single performer might eventually record an entire piece without 
making a single mistake, even that is difficult; you can imagine, then, 
the likelihood of getting a perfect performance from a group in one 
take. The likelihood diminishes rapidly as the number of players in 
the group goes up. Therefore, in order to produce the best possible 
finished recording, a number of intercuts — splices of tape taken from 
several different recorded takes — may ultimately be joined to produce 
the finished piece. You can use this technique as effectively as pro- 
fessional recording engineers; it simply requires a bit of basic infor- 
mation about splicing, accompanied by a fair amount of practice. 

As is implied by the foregoing discussion, splicing is the art of cut- 
ting the recording tape and then rejoining two separate pieces. The 
cutting should be done with a tape splicer. Such a unit will provide 
the tape cut at the correct angle, illustrated in Figure 5.10. The splice 
should always be made diagonally in order to avoid getting a "pop" 
in the recording at the point where the splice is made. In addition, 
you should be sure to use actual splicing tape — not just any kind 
of transparent tape. Not only will most transparent tape eventually 
slip and/or stretch, but the adhesive backing will sometimes ooze out 
and come into contact with delicate parts such as the tape heads. 

Where to splice is the part that will require practice. A good tape 
editor is an artist — an artist who is able to catch a singer between 
breaths, or a rock group between beats — and when the two segments 
are spliced together, you cannot tell that the material was not done 
in one take. To find the correct spot to splice, listen as the tape runs 
until it comes to the part you want to cut out. As the tape reaches 
that portion of the take, press the "pause" or "edit" button of your 


machine. This should leave the heads in the playback configuration, 
permitting you to turn the reels back and forth by hand until you are 
able to determine accurately the precise spot where you want to make 
the first cut. Repeat the procedure to find the spot for the second 
cut. This process will be made easier by removing the protective cover 
over the heads. In this manner, you will be able to actually see the 
playback head, and determine more accurately where the splice should 
be made. 


Figure 5.10. Correct tape-splice angle. 

When you have located the spots, mark them on the tape with a 
wax pencil or crayon. Be careful not to get wax on the heads! Then 
gently pull the tape away from the heads, so that you have a semicircu- 
lar loop including both marks between your right and your left hands. 
Put this loop in the splicer, positioning the first mark so that the first 
cut will be made at that point. Do the same for the second. After 
making the second cut but before moving either end of the recording 
tape, secure your splice with the special splicing tape. Assuming that 
all of the foregoing proceeds smoothly, you should then move the 
reels a sufficient amount to take up the slack in the tape and then 
rewind to a point just before the splice. This can usually be done by 
hand-turning the feed reel in a counterclockwise direction. Then dis- 
engage the pause control and listen to your splice. If it's a good one, 
you won't hear any superfluous noise, nor will the rhythm skip a 
beat or the singer miss a word. It should sound as if the original per- 
formance had continued perfectly, without interruption. 


As you can imagine, this is not so easy at first, but it is not really 
that difficult and is extremely rewarding once mastered. One obvious 
hint: such splices will be easier to make if the material is recorded at 
the highest possible tape speed. Whereas a recording made at 1 7/8 ips 
would require extremely careful editing, the tape only having moved 
less than two inches in one second, a recording made at 7V6 ips will 
allow some latitude for less-than-perfect cuts — four times the amount 
of tape will have passed over the record head in the same period of 
time. For this reason, most professional studios record at no less 
than 15 ips, and never at less than IV2. These speeds also have the 
residual effect (particularly on less expensive machines) of eliminating 
the unintended waver, or vibratolike effect, created by a variance in 
the tape speed as it moves across the heads. 

Experiment 1: Editing Techniques 

Before going on, practice making an intercut. This experiment can 
be performed with either a monaural machine or a four-channel. 
Record a simple melody, such as "Row, Row Your Boat" or "Merrily 
We Roll Along." Play the verse three times through. Then, to practice 
editing, go back and cut out the second verse, joining the first to the 
third. To do this effectively, you will now have to be concerned not 
only with editing techniques but also with the recording considerations 
involved. The third verse must be at the same tempo as the first; if 
you have not maintained the same tempo, the chances are that the 
difference between verses 1 and 3, when verse 2 is edited out, will be 
immediately noticeable. You can see that this is a particular problem 
when splicing material together from several different takes which may 
not have been made on the same day. 

You should also be careful to watch the volume level when record- 
ing. If verse 3 is significantly louder than verse 1, the abrupt change in 
volume where they are spliced together provides a sure tip to the 
listener that a splice has been made. Again, this concern is even more 
critical if you will be splicing pieces of cuts made over a series of days, 
or even weeks. Watch the level meters (V.U. meters) of your recorder 
carefully; try to maintain a consistent practice when recording so that 
the general levels of any two sessions will always work out to be 
pretty much the same. 

Finally, you must be attentive to the general acoustical properties 
of any two cuts that you are splicing together. The chances are that, 
in this experiment, verses 1 and 3 will not differ significantly — unless 


you deliberately add reverb or change the basic timbre of the sound 
during verse two. When editing takes from several sessions, however, 
you will not enjoy this luxury; just as you cannot splice together one 
cut made in your living room with a second cut made in your bath- 
room, you cannot splice together two different cuts exhibiting differ- 
ent amounts of electronic reverberation. It would be as obvious as 
splicing the "live," reverberation-saturated cut made in a bathroom 
together with the muffled, deadened cut recorded in a heavily car- 
peted, well-draped living room. 

Tape Manipulation Techniques 

Thus far, we have dealt with fairly common, though extremely 
useful, recording techniques. As a conclusion to this section of the 
text, a number of more exotic tape techniques will be presented. 
While the Odyssey alone permits almost total control of every para- 
meter of sound, the use of a tape recorder will extend your control 
just that much further, expanding your creative and technical 

One such technique is the relatively simple process of recording a 
track at one speed for playback or mixdown at another speed. This 
technique is most useful when you are recording a musical passage 
that is difficult for you to play at the speed required. To achieve 
the effect that you want, simply play the passage one octave lower 
than written and at half the speed you are ultimately seeking. If you 
record this octave-lower/half -speed passage at 3%, by playing it back 
at VA you will exactly double the speed and raise the pitch by 
one octave — producing exactly the effect you sought. Similarly, 
the same effect would be created by recording it at Th and playing 
it back at 15 ips. While it is less useful musically, you can reverse the 
procedure to create a number of interesting effects. A cymbal crash 
at half-speed, for example, is an effect that can be incorporated into 
certain electronic compositions. 

The same basic technique can be used to record and modify natural 
sounds, such as voices, footsteps, or running water. The act of record- 
ing, modifying, and assembling such sounds is the basis for musique 
concrete (classical electronic music made by splicing together unrelated 
phrases, notes, and sounds to achieve new effects). You should not 
ignore the possibility of mixing sounds of this nature with the elec- 
tronic effects produced by the synthesizer in order to assemble a 
compositional whole. 


A second technique you may wish to try is the creation of a tape 
loop. To do this, set the Odyssey controls as shown in the patch that 
appears in Figure 5.11. This patch will provide a random, percussive 
effect. Record two or three minutes of this effect. 

After you have a length of tape containing nothing but the percus- 
sive sounds, cut a portion of this tape (about thirty inches long) from 
the center of the tape. You do not necessarily have to be concerned 
about where the cut made begins and ends. After removing this section 
of tape, splice its two ends together to form a continuous loop. If 
possible, you should now turn your tape machine on its side so that 
gravity will tend to hold the loop in position as it is played. Be certain 
that the dull side of the tape is touching the heads of the recorder, 
since this is the side that the sound is recorded upon. Let the tape 
hang over the edge of the table, as shown in Figure 5.12. 

Figure 5.12. Tape loop in recorder. 

The repeating pattern that is created by the loop will often be one 
of sufficient interest that it can be used as a background accompani- 
ment to an electronic composition. You should also try taking this 
one step further by constructing deliberately repetitive patterns — 
perhaps a bass pattern that can be used as a foundation for a particu- 
lar composition. Simply record the pattern you want; then edit it 
carefully from the main reel, leaving an extra length of tape after the 
final beat of the pattern. This length of tape will affect the meter of 
the loop when the ends are joined. It's better to start with too much 
tape there and cut some out than to leave yourself too little and be 
forced to splice in minute sections until the timing is just right. 



An additional technique that you may wish to try is that of back- 
ward recording. This technique has been used with great success in 
the albums of a number of contemporary groups. The attraction of 
the technique is that it totally reverses the attack and decay character- 
istics of every sound. This effect is particularly noticeable when re- 
cording percussion instruments; try a rhythmic series of cymbal 
strokes. Record the pattern just as the drummer plays it. Then flip 
the reels over as illustrated in Figure 5.13. 

Remember that you have now turned the tape upside down; 
therefore, on a four-channel machine, the track order will now be 
reversed: track 1 will be track 4, track 2 will be track 3, track 3 will 
have become track 2, and track 4 becomes the new track. Conse- 
quently, this technique will not work on quarter-track stereo recorders. 
With the cymbal track now playing backward, you can add other tracks 
to this unusual percussive background. The result is one track playing 
backward, the others forward, both at the same time and perfectly 

Figure 5.13. Reversing tape reels to record backward. 


Other effects, such as the deliberate use of the time delay produced 
by the separation of the record and payback heads, can be created 
on certain machines. The principle of this effect, of course, is to feed 
the signal from the output of a particular channel (taken from the play- 
back head) back into the input for the same channel, thus creating 
an echo effect by returning the signal produced by the playback 
head to the record head. Consult the owner's manual for your machine 
to determine if this is possible; a number of factors that vary widely 
from one machine to another will have a pronounced effect upon the 
quality and duration of the echo. These factors include: (1) the dis- 
tance between the record and playback heads, (2) tape speed, and 
(3) the level of the signal. The louder the signal, the slower the tape 
speed; the wider the distance between the two heads, the longer the 
echo. If your machine can be operated in this way, experiment to 
see if musically useful results can be produced. 

Conclusion: Listening for Electronic Music 

While it has been the primary purpose of this text to get you invol- 
ved in the actual creation of electronic music, it is hoped that a 
secondary goal has also been achieved: that of making you more aware 
of and more curious about the sounds you hear around you every day. 
Every natural sound could be synthesized, given the proper electronic 
equipment. Imagine how you might create patches for many of the 
sounds you hear during your daily activities; listen for other sounds 
that might be recorded and then modified for use in an electronic 
composition. Be alert to your audio-environment! 

Your awareness will be rewarded by an ever-increasing amount of 
good electronic music — music being played by the top rock, jazz, 
and even classical artists. Nearly every performer of major stature who 
plays a keyboard instrument is now using a synthesizer, either in live 
performance or in his records. Moreover, the use of such instruments 
extends beyond all limiting classifications; musicians performing all 
kinds of music have realized the potentially expanded capacity for 
expression that electronic music synthesis can provide. Even television 
commercials, station identification themes, and the drama sound- 
tracks are incorporating an increasing amount of pure electronic 
effects into their audio makeup. 

It's not a fad, and it's not an esoteric fancy intended for a chosen 
few. Instead, electronic music should be regarded as what it is — the 
logical continuation in the evolutionary development of musical ex- 
pression. Those of us who have already experienced, even in a limited 
way, the boundless potential of this medium are convinced that in 
these possibilities lie the beginnings of an exciting and significant 
evolutionary step in the development of the music of the world. 
Enjoy it. 



Additive synthesis: adding sine waveforms together to create new 

Amplifier: an electronic circuit which increases the power of an 
electrical signal. 

Amplitude: amount of a waveform's deviation from center. When 
used to describe sound, amplitude means volume. 

Amplitude modulation: a periodic change in the amplitude of a 
sound; for instance, tremolo. 

Aperiodic waveform: irregular, nonrepeating waveform. 

Attack: beginning of a sound. 

Attenuator: controls amount of signal passing through it. 

Audio range: range of pitches you can hear.* Roughly 20-20,000 Hz. 

Band-pass filter: passes one frequency band. 

Band-reject filter: rejects one frequency band. 

Cutoff frequency: used when describing the characteristics of high- 
pass or low-pass filters to indicate the specific frequency 
beyond which the filter is supposed to attenuate all frequencies. 

Decay: initial fading of sound (after attack). 

Envelope: attack and decay of a sound. 

Envelope generator: produces transient voltages useful in creating 
attacks and decays and special effects. 

Filter: changes tone color (timbre) by removing selected harmonics. 

Frequency: rate at which a waveform repeats. Expressed in cycles 
per second or Hertz (Hz). 

Frequency modulation: a periodic change in the pitch of a sound; 
for instance, vibrato. 

Fundamental: usually the lowest frequency component in simple 
waveforms, perceived as the "pitch" of the sound. 

Gate: on/off signal indicating beginning, duration, and end of an event. 

Harmonics: overtones that give a tone a particular sound or timbre. 
Harmonic frequencies are always exact multiples of the 


Harmony: two or more simultaneous tones, implying a tonality. 
Hertz (Hz): term for cycles per second. 

High-pass filter: passes high frequencies, cuts out low frequencies. 
Low-pass filter: passes low frequencies, cuts out high frequencies. 

Low-frequency oscillator: an oscillator which is designed specifically 
to operate at subsonic frequencies. 

Mixer: combines signals. 

Modulation: any periodic change in a waveform. 

Noise: random signals which contain all audio frequencies. 

Oscillator: generates tone or low-frequency periodic waveform. 

Overtones: frequency components of a sound. May be in any 
mathematical relationship to the fundamental. 

Patch: connection of two or more functions. 

Periodic waveform: repeating wave pattern. 

Phase: relationship between waveforms at any moment in time. 

Phase-synchronization: forcing a fixed phase relationship between 
two waveforms. 

Pink noise: noise which is musically balanced; high and low 
frequencies sound equally loud. 

Pitch: perceived frequency of a sound. 

Pitch bend: changing the frequency of a pitch while played. 
Portamento: sliding between notes. 
Pulse wave: family of waveforms with square corners. 
Release: ending of a signal. 

Resonance; amplifies a band of overtones. A resonance which 
moves in frequency can create a "wow" sound. 

Ring modulator: produces a complex output from two simple input 

Rolloff : the effectiveness with which a filter eliminates signals which 
it is not supposed to pass. 

Sample and hold: a circuit which can be used to store, or hold, an 
input voltage. 

Sawtooth wave: sounds rich, full, brassy. 

Signature: signs placed at the beginning of a composition indicating 
the key. 


Sine wave: sounds smooth and pure; has no harmonics. 

Square wave: sounds hollow and reedy. Square wave is a special kind 
of pulse wave. 

Subsonic: below human hearing range; usually lower than 20 Hz. 

Subtractive synthesis: filtering out certain frequencies to create new 

Sustain: in synthesis, describes the level of the held part of a note in 
relation to the attack. 

Tremolo: amplitude modulation. 
Tempo: speed. 

Timbre: all those qualities of a sound that make it distinctive. 

Tonality: relationship of the pitches defining a "key." 

Transposition: changing from one key signature to another. 

Trigger: electronic impulse used most often to activate envelope 

Vibrato: frequency modulation. 
Voltage: electrical potential. 

Voltage control: a process whereby one electrical circuit is used to 
control the function of some other electrical circuit. 

Voltage-controlled amplifier: an amplifier whose gain can be con- 
trolled by an external voltage. 

Voltage-controlled filter: a filter whose cutoff frequency can be con- 
trolled by an external voltage. 

Voltage-controlled oscillator: an oscillator whose operating frequency 

can be controlled by an external voltage. 
Waveform: characteristic shape of a wave; helps determine timbre. 



ADSR. See Envelope Generator, 

Amplifier. See Voltage-Controlled 

Amplifier (VCA). 
Amplitude: 5-11 

Amplitude modulation: 23-24, 142. 

AR. See Envelope Generator. 

Attenuators ): 29-31. 

Audio Mixer: 36. 

Beats: 165-67. 

Brightness. See Timbre. 

Cycle: 2, 4, 7. See also Hertz (Hz). 

Decay. See Envelope. 

Ear training: 164-67. 

Envelope: 18-21, 93-94; attack, 

18-20, 94-95, 132-35, 160; 

decay, 18-20, 94, 132, 134-35, 

160; release, 21, 94; sustain, 21, 94. 
Envelope Generator: 19-24; ADSR, 

29, 64-66, 85-91, 93-101, 136, 163; 

AR, 29, 86-93, 95-96. 
Events: 17-19,40. 
Filtering: 12-16, 20-21, 35, 70, 

169-74. See also Synthesis, 

additive vs. subtractive. 
Filter(s): band-pass, 15, 169 (effect); 

band-reject, 15; high-pass (HPF) , 

14, 60, 68-70, 169, 173; low-pass, 

13-14, 16, 169, 172. 
Filter, Voltage-Controlled: 20, 60-66, 

72, 83, 88-90,93, 107, 126, 

135-36, 160; used as VCO, 72-75. 
Foot-pedal controllers: 126-27. 
Frequency : 7-9, 38; audio, 17; 

subsonic (low), 7, 17-19, 40. See 

also Harmonics. 
Frequency modulation (FM): 23-24, 

51-55, 142. 
Harmonics: 10-12, 16, 132, 140-42, 

168-74; listen to, 169, 172. 
Harmony: 131, 164-67; chords, 176-77. 
Hertz (Hz): 7. See also Cycle; 

HPF. See Filtering; Filter(s). 
Keyboard: 22, 23, 42, 74-75, 116-28; 

polyphonic, 128; intervals, 124, 


(Keyboard, continued) 

165-67. See also Voltage control. 

Low-Frequency Oscillator (LFO): 
51, 66, 100. 

Melody: 155-63, 176-77; counter- 
melody, 175. 

Modulation. See Amplitude modulation; 
Frequency modulation. 

Music studio, electronic: 131, 187-200. 

Noise: pink, 32-35; white, 29, 32-35, 

Noise Generator: 29, 30. 
Orchestration: 163. See also Trans- 

Oscillator. See Low-Frequency Oscil - 
lator; Voltage-Controlled Oscillator. 
Overtones. See Harmonics. 
Patch: 27, 31. 

Pitch: 2-4, 7, 9, 17-18. See also 

Pitch bend: 122-23. 
Portamento: control, 117-19, 126. 
Pulse width: 46-48. 
Pulse-width modulation: 49-50,97, 

Release. See Envelope. 
Resonance: 13-14, 62-64. See also 

Filter, Voltage-Controlled. 
Rhythm: 155-57. 
Ring Modulator: 76-79. 
Rolloff: 13-14. See also Filter(s). 
Sample and Hold (S/H): 108-15. 
Shape. See Envelope. 
Signal modifier: 31, 60, 68. See also 

Filter, Voltage-Controlled; Filter(s), 

high-pass; Ring Modulator; Voltage - 

Controlled Amplifier (VCA). 
Signal source: 31, 60. See also Noise 

Generator; Voltage-Controlled 

Slide controls: 29-30. 
Sustain. See Envelope. 

Synchronization: 55-60. See also 
Voltage-Controlled Oscillator. 

Synthesis: additive vs. subtractive, 
12-15. See also Filtering. 

Tape music techniques: 187. See also 
Tape recorder, editing techniques, 
multitrack recording. 

Tape recorder: 187-200; editing tech- 
niques, 201-208; four-channel, 
194-96; monaural, 191-92; 
multitrack recording, 192-93, 
196-200; stereo, 192-194. 

Tempo: 156-57. See also Rhythm. 

Timbre: 1-2, 8-9, 132-42, 155, 160-62; 
modulation of, 66. 

Tone generator. See Filter, Voltage- 
Controlled, used as VCO; Voltage- 
Controlled Oscillator. 

Tonality: 155, 158-60. See also Pitch. 

Transpose switch: 42,44,120-21. 

Transposition: 178-86; notation, 187. 

Tremolo: 66, 142. See also Timbre, 
modulation of. 

Trill: 51, 104. 

Tuning. See Keyboard, intervals; 

Synchronization; Ring Modulator. 
Voltage-Controlled Amplifier (VCA): 

22, 60; control of, 80-90, 93-95. 
Voltage-Controlled Filter (VCF). See 

Filter, Voltage-Controlled. 
Voltage-Controlled Oscillator (VCO): 

22-23, 32, 36-49, 54-60, 128. 
Vibrato: 23, 51, 103, 142. 
Voltage control: 20-23. 
Volume: 5-11. 

Waveforms: definition, 2-5; aperiodic, 
4-5, 35-36; harmonics of, 11-12; 
periodic, 4-5. See also Events; 
Filter, Voltage-Controlled; Low- 
Frequency Osicllator; Voltage- 
Controlled Oscillator. 

Waveshape. See Waveforms.